KR20100121389A - Storage device based on a flash memory and user device including the same - Google Patents

Abstract

PURPOSE: A storage unit based on a flash memory preventing exposure of secure data and a user device including the same are provided to improve the response speed about a host command. CONSTITUTION: A host transmits a invalidation command to a storage device(S1100). If the capacity of a area to delete exceeds the reference capacity, a SSD(Solid State Drive) controller records a location of the area to delete(S1110,S1120). The SSD controller processes the data recorded in the area to delete to a invalid data without recording the location of the area to delete(S1140). The SSD controller notifies the completion of execution of the invalidation command to the host(S1130).

Description

STORAGE DEVICE BASED ON A FLASH MEMORY AND USER DEVICE INCLUDING THE SAME}

The present invention relates to a storage device based on a flash memory and a user device including the same.

User devices are not only hosts such as personal computers, digital cameras, camcorders, mobile phones, MP3, PMPs, PDAs, etc., but also memory cards, USB memories, solid state drives (SSDs), hard disk drives (HDDs), etc. It may include a storage device. The user device may include a storage device inside the host. The storage device may include volatile memory such as DRAM, SRAM, and the like, and nonvolatile memory such as EEPROM, FRAM, PRAM, MRAM, and Flash Memory.

Data stored in the storage device may be deleted by a user's command. Generally, the delete operation is performed through the host's file system (for example, FAT). The file system issues a delete command to the storage device. In addition, the file system can recover deleted data under certain conditions. The file system authorizes a recovery command to recover the data.

An object of the present invention is to provide a storage device and a user device including the same that can improve the response speed to the host command. Another object of the present invention is to provide a storage device capable of preventing the exposure of secure data and a user device including the same.

A user device according to the present invention includes a storage medium for storing data; And a controller for controlling the storage medium, wherein the controller may selectively perform an invalidation operation or a logging operation according to whether the size of data to be deleted exceeds a reference size. The reference size is variable through a firmware update.

In an embodiment, the user device may further include a host that manages file data using a file system, and the controller may perform invalidation using the file system information of the host. The storage medium may include an area for storing the file system information.

In another embodiment, a user device includes a host for providing an invalidation command; And a buffer memory for temporarily storing data to be stored in the storage medium, wherein the controller may invalidate data stored in the buffer memory in response to the invalidation command. The controller may include a mapping table for invalidating data stored in the buffer memory. The controller may update the mapping table in response to the invalidation command.

In another embodiment, the user device may further include a host having a buffer memory for temporarily storing data to be stored in the storage medium, and the host may perform an invalidation operation on the buffer memory. The host may include a mapping table for invalidating data stored in the buffer memory.

As another embodiment, when a write operation is requested, the controller may determine whether the write operation is related to metadata and update the mapping information of the user data according to the determination result. When it is determined that the write operation is related to partition metadata, the controller may update the mapping information so that user data corresponding to the partition metadata is invalidated.

As another example, the controller of the user device may delay an erase operation on data to be deleted. The controller may include an invalid delay queue for delaying an invalidation operation on data to be deleted. Data to be deleted may be sequentially registered in the invalid delay queue, and may be sequentially invalidated according to a first-in first-out method.

Another aspect of the user device according to an embodiment of the present invention is a storage medium for storing data; And a controller for controlling the storage medium, wherein the controller may determine whether the data to be deleted is data that requires security, and selectively perform an invalidation operation according to the determination result. Here, when the data to be deleted does not require security, the controller may selectively perform an invalidation operation or a logging operation according to whether the size of the data to be deleted exceeds a reference size. When the data to be deleted requires security, the controller can perform an invalidation operation without delay.

In an embodiment, the user device may further include a host that manages file data using a file system, and the controller may perform invalidation using the file system information of the host.

In another embodiment, a user device includes a host for providing an invalidation command; And a buffer memory for temporarily storing data to be stored in the storage medium, wherein the controller may invalidate data stored in the buffer memory in response to the invalidation command.

As another embodiment, when a write operation is requested, the controller may determine whether the write operation is related to metadata and update the mapping information of the user data according to the determination result.

In another embodiment, the controller may include an invalid delay queue for delaying an invalidation operation on data to be deleted.

According to the present invention, the response speed to the host command can be improved, and another object of the present invention can be prevented from exposing security data.

DETAILED DESCRIPTION Hereinafter, exemplary embodiments of the present invention will be described with reference to the accompanying drawings so that those skilled in the art may easily implement the technical mapping of the present invention. .

A user device includes a host and a storage device. The user device may include a processor, main memory, and a storage medium in hardware. Here, the processor or the main memory may be included in the host, and the storage medium (eg, flash memory or PRMA, etc.) may be included in the storage device. In addition, the user device may be software, and may include a user application, an O / S, a file system, a flash translation layer (FTL), and the like.

File (s) created on the host are managed by the file system. Under the control of the host's processor (eg, CPU), the file data is transferred to the storage device. The transferred data will be stored in a storage medium of the storage device (eg flash memory or PRAM, etc.). Meanwhile, file data requested by the host will be transferred from the storage device to the host's main memory.

If any file is deleted from the host, the file system will treat the file as deleted. Deleting a file means that the metadata of the deleted file is changed as file system management data. Even if a file is deleted from the host (in other words, even if the deleted file's metadata is changed by the file system), the storage device cannot determine whether the file is invalid.

Therefore, a merge operation or garbage collection operation is performed on the invalid file, which means that the performance of the storage device is degraded. In addition, the invalid file is stored in the storage device as if it is valid data, which means that the effective storage space of the storage device is reduced.

The storage device or the user device including the same according to an exemplary embodiment of the present invention may reduce a merge operation or garbage collection operation on invalid data through an invalidation operation. According to the present invention, the operating performance of the storage device or the user device can be improved, and the storage space and the service life of the storage device can be increased. Hereinafter, various embodiments of a user device performing an invalidation operation will be described.

Ⅰ. Logging  User device that performs invalidation using

1 is a block diagram illustrating an example of a user device performing an invalidation operation using a logging scheme. Referring to FIG. 1, a user device 1000 according to an embodiment of the present invention includes a host 1100 and a storage device 1200.

The host 1100 may be configured to control the storage device 1200. The host 1100 includes, for example, a portable electronic device such as a personal / portable computer, PDA, PMP, MP3 player, or the like. When the contents (eg, files) of data stored in the storage device 1200 are deleted from the host 1100, the host 1100 processes metadata related to a file of the deleted contents, thereby removing the contents of the deleted contents. Invalidate the file.

The host 1100 notifies the storage device 1200 of invalidation (or deletion) of files as necessary. This may be accomplished by the host 1100 sending a specific command to the storage device 1200. Hereinafter, such a specific command is referred to as an "invalidity command". The invalidation command may include information for specifying an area to be deleted (for example, address information). The invalidation command may be used in various names such as a file delete command, a trimity command, an unwrite command, and a deletion command.

Processing of the metadata for the file to be deleted will be done by the file system of the host 1100. The file system will delete only the file's metadata, not the file's contents, for faster operation. As an example of the FAT file system in the file system, one special code may be used as the first character of a file name to indicate a deleted file. For example, place the hexadecimal byte code 'E5h' in place of the first character of the file name. This is a special tag that says 'This file has been deleted'.

When any file is deleted, the file system will place one special code in the first character of the file name of the deleted file. When the metadata of a file to be deleted is processed by the file system, the content of the file to be deleted is treated as invalid data in the file system, while the storage device 1200 remains valid data.

For this reason, the storage device 1200 will recognize the memory block including the data of the deleted file as a valid block. Thus, operations such as a merge operation or a garbage collection operation are performed on the memory block (including the data of the deleted file), and as a result, the performance of the storage device 1200 or the user device will be degraded. To prevent this, the file system of the host 1100 may provide an invalidation command to the storage device 1200 such that the contents of the deleted file are substantially invalidated in the storage device 1200.

1, the storage device 1200 may operate under the control of the host 1100. The storage device 1200 will retain the stored data even when power is cut off. The storage device 1200 may be, for example, a solid-state drive (SSD). However, it will be appreciated that the storage device 1200 is not limited to SSD.

The storage device 1200 may include an SSD controller 1220A and a storage medium 1240. The SSD controller 1220A may control the storage medium 1240 in response to a request from the host 1100. The SSD controller 1220A may be connected to the storage medium 1240 through a plurality of channels CH1 to CHn. The storage medium 1240 may be composed of a plurality of nonvolatile memory chips. A plurality of nonvolatile memory chips may be commonly connected to each of the channels CH1 to CHn.

The storage medium 1240 may be composed of a plurality of flash memory chips, for example. However, the storage medium 1240 may be configured with other nonvolatile memory chips (eg, PRAM, FRAM, MRAM, etc.) instead of flash memory chips. Each of the flash memory chips constituting the storage medium 1240 may store 1-bit data or M-bit data (M is an integer of 2 or greater) per cell.

According to the user device 1000 according to an embodiment of the present disclosure, when an invalidation command is provided from the host 1100, the storage device 1200 records / records only a location of an area (or files) to be deleted according to the invalidation command. And notifies the host 1100 that execution of the requested command has been completed.

In other words, the storage device 1200 will not just invalidate the area of the files to be deleted, but will record / store only the location of the area of the invalid files. This will be explained in detail below. In this case, it is possible to process the request of the host 1100 quickly. That is, the response time for the request of the host 1100 may be shortened. In addition, it is possible to improve the performance of the storage device 1200 or the user device 1000 through a quick response.

FIG. 2 is a block diagram schematically illustrating the SSD controller shown in FIG. 1. Referring to FIG. 2, the SSD controller 1220A may include a host interface 1222, a flash interface 1224, a processing unit 1226, and a buffer memory 1228. It will be appreciated that the components of SSD controller 1220A are not limited to those disclosed herein. For example, it will be appreciated that the SSD controller 1220A further includes an error correction block for detecting and correcting errors in data stored in the storage medium 1240.

 The host interface 1222 will provide an interface with the host 1100, and the flash interface 1224 will provide an interface with the storage medium 1240. The processing unit 1226 generally controls the operation of the SSD controller 1220A, and the buffer memory 1228 will be used to temporarily store data to be stored in or read from the storage medium 1240. In addition, the buffer memory 1228 will be used to store the above-described history (writing of the area to be erased according to the invalidation command).

In an exemplary embodiment, an additional volatile / nonvolatile memory (or register) may be used instead of the buffer memory 1228 to store the aforementioned history. However, it will be understood that the medium for recording the position of the area to be deleted in accordance with the invalidation command is not limited to that disclosed herein.

3 is a block diagram illustrating another embodiment of the SSD controller shown in FIG. 1. Referring to FIG. 3, the SSD controller 1220B may include a host interface 1222, a flash interface 1224, a plurality of processing units 1226_1 to 1226_N, and a buffer memory 1228. It will be appreciated that the components of SSD controller 1220B are not limited to those disclosed herein. For example, it will be appreciated that the SSD controller 1220B further includes an error correction block for detecting and correcting errors in data stored in the storage medium 1240.

The SSD controller 1220B may perform various operations through the plurality of processing units 1226_1 to 1226_N. The SSD controller 1220B may divide various control operations and then allocate control operations corresponding to each processing unit. Even if the SSD controller 1220B is driven by a low frequency clock, the performance of the SSD controller 1220B including the plurality of processing units 1226_1 to 1226_N may be improved.

4 is a flowchart illustrating an operation of a user device according to an exemplary embodiment of the present invention. Hereinafter, the operation of the user device according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

The host 1100 may provide an invalidation command to the storage device 1200 as needed. For example, the invalidation command may be used to delete files having invalid contents among files stored in the storage device 1200.

First, in operation S1100, an invalidation command will be transmitted from the host 1100 to the storage device 1200. As mentioned above, the invalidation command will include information (e.g., address information) indicating the area of the files with invalid content.

In operation S1110, the storage device 1200, that is, the SSD controller 1220 may determine whether the capacity of the region to be deleted exceeds the reference capacity according to the invalidation command. If it is determined that the capacity of the area to be deleted exceeds the reference capacity, in step S1120, the SSD controller 1220 may record / store the location of the area to be deleted. This operation is called "logging". After the location of the region to be deleted is recorded, in step S1130, the SSD controller 1220 will notify the host 1100 that execution of the invalidation command is completed. The procedure will then end.

In an embodiment, the SSD controller 1220A includes a buffer memory (for example, DRAM), and information indicating the location of an area to be deleted will be recorded in the buffer memory. The position of the region to be erased can be recorded using a bitmap structure indicating the region to be erased where each bit exceeds the reference capacity. However, it will be understood that the logging method is not limited to the method of recording the location of the area to be deleted. For example, it is possible to record the address information of the area to be deleted.

The history in which the location of the region to be deleted is recorded may be stored in the storage medium 1240 at a given time or as needed. The stored history will be loaded into SSD controller 1220 at power-on. On the contrary, the history in which the position of the area to be deleted is recorded can be managed using the buffer memory. When the idle time occurs, the SSD controller 1220 invalidates the area to be deleted by using the history stored in the buffer memory or the storage medium 1240.

In this case, the invalidation means processing the data recorded in the area to be deleted as invalid data. This invalidation can be done by managing a mapping table in which the mapping between the physical and logical blocks of the storage medium 1240 is recorded. For example, such invalidation can be done by mapping out mapping information for the area to be deleted from the mapping table or by marking the area to be deleted in the mapping table. However, it will be understood that invalidation is not limited to what is disclosed herein.

The flash memory chips constituting the storage medium 1240 are managed by a flash translation layer (FTL) executed by the SSD controller 1220. Management of the mapping table will be done by such FTL, for example. Invalidated areas of the storage medium 1240, that is, invalidated memory blocks, will be erased as needed.

In an embodiment, the reference capacity will be set to be variable in hardware or software. For example, the reference capacity may be set to be variable through an update of the firmware (FTL) of the storage device 1200. Alternatively, the reference dose may be set to be variable by the host 1100. In such a case, setting the reference capacity by storing a specific value representing the reference capacity in a register of the host interface 1222 (used to store the reference capacity) during the recognition procedure between the host 1100 and the storage device 1200 may be necessary. It will be possible. The region to be deleted represents a logical region and will be converted to the physical region of the storage medium 1240 by FTL.

According to an embodiment, the capacity of the region to be deleted according to the invalidation command may be limited according to the storage device 1200. In this case, the maximum capacity of the area to be deleted is recorded in the storage device 1200 through an invalidation command, and the host 1100 uses the information indicating the maximum capacity of the area to be deleted recorded in the storage device 1200 to invalidate the command. Will cause.

Returning to step S1110, if it is determined that the capacity of the area to be deleted does not exceed the reference capacity, the procedure will proceed to step S1140. In operation S1140, the SSD controller 1220 may immediately process data written in the region to be deleted as invalid data without recording the position of the region to be deleted. This invalidation can be done by mapping out the mapping information for the region to be deleted from the mapping table, or by marking the region to be deleted in the mapping table, as mentioned above.

After immediately processing the data recorded in the area to be deleted as invalid data, in step S1130, the SSD controller 1220 will notify the host 1100 that execution of the invalidation command is completed. The procedure will then end.

The storage device 1200 according to an exemplary embodiment of the present invention may execute an invalidation command for a given / limited time (for example, a command from the host) regardless of the capacity of the region to be deleted through the aforementioned logging scheme. Time to process). It is also possible to prevent the program and erase operations for the invalidated area from being performed unnecessarily.

The storage device 1200 may be implemented to perform a background operation during the extra time (difference between the given / limited time and the substantial instruction processing time) resulting from quickly processing the command within the given / limited time of the host 1100. In this case, a response to the requested command will be made before the given / limited time has passed.

When the storage device 1200 is composed of flash memory chips, even if data stored in a specific area of the storage medium 1240 is invalidated at the request of the host 1100 (eg, an invalidation command), the storage medium 1240 The data stored in a specific area of) remains substantially intact due to the nature of the storage medium 1240 which does not support overwriting. This is because the physical area of the storage medium 1240 is not substantially managed, but only mapping information is managed through the FTL.

Storage device 1200 may be used in devices (eg, printers) that require security. In this case, after the data requiring security has been processed, such data will remain in storage 1200. This means that secure data can be exposed unintentionally. In the user device according to an embodiment of the present invention, when the data requiring security is deleted, the host 1100 may store a security invalidation command or an invalidation command including information indicating deletion of the security data. Will provide). This will be described in detail later with reference to FIG. 5.

5 is a flowchart illustrating an operation of a user device according to another exemplary embodiment. Hereinafter, the operation of the user device according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

The host 1100 may provide an invalidation command to the storage device 1200 as needed. The invalidation command may be classified according to general data and data requiring security (hereinafter referred to as security data). For example, the host 1100 may provide an invalidation command to the storage device 1200 when requesting deletion of general data and a security invalidity command when requesting deletion of security data. Alternatively, the host 1100 may provide the storage device 1200 with an invalidation command including information indicating deletion of general data or deletion of security data.

First, in operation S1200, a command may be transmitted from the host 1100 to the storage device 1200. For convenience of explanation, assume that a command for notifying an area to be deleted is input. If such a command and other command are input, the corresponding operation will be performed. In operation S1210, the storage device 1200, that is, the SSD controller 1220 may determine whether the input command is an invalidation command or a security invalidation command. As another example, the SSD controller 1220 will determine whether the particular bit (s) of the invalidation command entered indicates deletion for general data or deletion for security data.

If it is determined that the input command is an invalidation command, the procedure proceeds to S1220. In operation S1220, the SSD controller 1220 may determine whether the capacity of the region to be deleted exceeds the reference capacity according to the invalidation command. If it is determined that the capacity of the area to be deleted exceeds the reference capacity, in step S1230, the SSD controller 1220 may record / store the location of the area to be deleted in the buffer memory. After the location of the area to be deleted is recorded, the procedure will proceed to step S1250.

If it is determined that the capacity of the region to be deleted does not exceed the reference capacity, the procedure will proceed to step S1240. In operation S1240, the SSD controller 1220 may immediately process data written in the region to be deleted as invalid data without recording the position of the region to be deleted. This invalidation can be done by mapping out the mapping information for the region to be deleted from the mapping table, or by marking the region to be deleted in the mapping table, as mentioned above.

After immediately processing the data recorded in the area to be deleted as invalid data, the procedure will proceed to step S1250. In operation S1250, the SSD controller 1220 may notify the host 1100 that execution of the invalidation command is completed. The procedure will then end.

Returning to step S1210 again, if it is determined that the entered command is a security invalidation command, the procedure will proceed to S1260. In step S1260, the SSD controller 1220 will immediately process the data recorded in the area to be deleted as invalid data. In addition, the SSD controller 1220 may control the storage medium 1240 such that the memory block (s) of the storage medium 1240 corresponding to the area to be erased are substantially erased. That is, secure data stored in the storage medium 1240 may be erased (or lost / removed). Thereafter, in step S1250, the SSD controller 1220 will notify the host 1100 that execution of the input command is completed. The procedure will then end.

The storage device 1200 according to an exemplary embodiment of the present invention may execute an invalidation command for a given / limited time (for example, a command from the host) regardless of the capacity of the region to be deleted through the aforementioned logging scheme. Time to process). In addition, it is possible to prevent the exposure of the secure data of the storage device 1200 by performing an erase operation on the storage medium 1240 without immediately / delaying when a command for deleting the secure data is input.

In the storage device 1200 according to an exemplary embodiment of the present disclosure, unlike the description of FIGS. 4 and 5, the capacity of the region to be deleted may be determined based on the invalid capacity according to the invalidation command. The position of the area to be deleted can be recorded / stored in the buffer memory without determining whether or not it is exceeded. In other words, the storage device 1200 may be implemented to write a location of an area to be erased to the buffer memory whenever an invalidation command is input. Invalidation of the logged area will be done during idle time.

In an embodiment, the invalidation described above may optionally involve an operation in which information indicating an invalid block is recorded in the memory block (s) of the storage medium 1240 corresponding to the area to be deleted. Alternatively, the invalidation described above may involve an operation in which information indicating an invalid block is recorded in the memory block (s) of the storage medium 1240 corresponding to the area to be deleted.

II. User device performing invalidation operation using file system information

6 is a block diagram illustrating an embodiment of a user device for performing an invalidation operation using file system information. The user device 2000 illustrated in FIG. 6 includes a host 2100 and a data storage device 2200.

From the standpoint of the host 2100, the data storage device 2200 is regarded as a storage medium free of read, write, and erase operations as a hard disk. The data storage device 2200 may include a flash memory 2210 and a controller 2220. Here, other nonvolatile memory (eg, PRAM, FRAM, MRAM, etc.) may be used instead of the flash memory 2210.

The flash memory 2210 is composed of a large number of memory cells having a string structure. This set of memory cells is called a cell array. The memory cell array of the flash memory 2210 is composed of a plurality of blocks. Each block consists of a plurality of pages. Each page consists of a plurality of memory cells that share one word line. The flash memory 2210 performs an erase operation in units of blocks and read and write operations in units of pages.

As described above, the flash memory 2210 has a different read / write unit and an erase unit. In addition, unlike other semiconductor memory devices, the flash memory 2210 may not be overwritten. Therefore, the flash memory 2210 always performs an erase operation before performing a write operation.

Due to these characteristics of the flash memory 2210, the data storage device 2200 needs to separately manage units of read, write, and erase operations in order to use the flash memory 2210 as a hard disk. Flash Translation Layer (FTL) is system software developed for this purpose.

Referring to FIG. 6, the flash memory 2210 includes a FAT area 2211, a data area 2212, a log area 2213, and a meta area 2214. Here, the FAT area 2211 is a kind of file system area.

In the FAT area 2211, file system information (e.g., FAT information) is stored. The log blocks of the log area 2213 are each assigned to some of the data blocks of the data blocks of the data area 2212. That is, when data is to be written to the data block of the data area 2212, the data is not directly written to the data block but stored in the corresponding log block of the log area 2213.

However, if a log block of the log area 2213 corresponding to the data block of the data area 2212 is not specified, or there is no empty page in the log block of the log area 2213, or a host ( If there is a request of 2100, the merge operation is performed. Through the merge operation, valid pages of the log block and valid pages of the data block are copied into a new data block or log block. When the merge operation is performed, the mapping information is changed, and the changed mapping information is stored in the meta area 2214.

The controller 2220 is configured to control the flash memory 2120 when there is an access request from the host 2100. As shown in FIG. 6, the controller 2220 includes a control logic 2221 and a work memory 2222. The flash memory layer FTL is stored in the work memory 2222. The control logic 2221 causes the flash translation layer FTL stored in the work memory 2222 to operate when there is an access request from the host 2100.

FIG. 7 is a conceptual diagram illustrating a merge method of the user device illustrated in FIG. 6. Referring to FIG. 7, the user device 200 performs an invalidation operation using file system information (eg, FAT information).

Referring to FIG. 6, valid pages 2511 and 2513 of log block 2510 and valid pages 2522 of data block 2520 are copied to a new data block 2530. The first and third pages 2511 and 2513 of the log block 2510 are copied to the first and third pages 2531 and 2533 of the new data block 2530, respectively. The second page 2522 of the data block 2520 is copied to the second page 2532 of the new data block 2530.

Here, the merge method of the data storage device according to the present invention selectively copies valid pages of the data block 2520 to a new data block 2530 with reference to the FAT information 2540.

Referring to FIG. 7, the FAT information 2540 stored in the flash memory 2210 stores information on whether the corresponding page of the data block 2520 is allocated. For example, since the first page 2521 of the data block 2520 is a page not used to store a file, it is indicated as NA (Not Allocation) in the FAT information 2540. The second page 2522 of the data block 2520 is a valid page for storing a file, and is indicated by A indicating allocating to the FAT information 2540. The FAT information 2540 corresponding to the third and fifth pages 2523 and 2525 of the data block 2520 is indicated by NA.

The fourth page 2524 of the data block 2520 is a valid page for storing a file, but is marked as D for deletion since it is an erased page from the file system point of view. That is, the data stored in the fourth page 2524 of the data block 2520 is a valid page from an FTL perspective, but is no longer valid from a file system perspective. Thus, if the fourth page 2524 of the data block 2520 has deleted data from a file system perspective, the data of that page is not copied to the new data block 2530.

Accordingly, the user device 2000 according to an embodiment of the present invention performs the invalidation operation by preventing the merge operation of data that is no longer valid with reference to file system information (for example, FAT information). According to the user device 2000 according to the present invention, the overall merge operation time may be reduced by not copying pages that are not necessary from the file system perspective.

FIG. 8 is a flowchart illustrating an invalidation operation of the user device illustrated in FIG. 6.

In operation S2610, the physical page of the new data block (see FIG. 7) 2530 is converted into a logical page. In step S2620, the FTL reads FAT information from the FAT area 2211 of the flash memory 22 (see FIG. 6). In operation S2630, it is determined whether the corresponding page of the data block is a valid page to be copied by referring to the FAT information. That is, it is determined whether the corresponding page of the FAT area is allocated for storing the file. If the file is a page that is not stored or a deleted page, the process proceeds to the next step (S2650) without copying to a new data block.

In operation S2640, if the corresponding page of the data block is a valid page in which the file is stored, the page is copied to a new data block. In step S2650, steps S2630 to S2640 are repeated for the remaining pages of the data block. After all the page copy of the data block is performed, the page copy of the log block is performed and the merge operation ends.

As described above, in the process of copying data, the user device 2000 according to an embodiment of the present invention refers to file system information (for example, FAT information) to copy invalid data from a file system perspective. Merge operation time can be reduced by preventing the The most time-consuming part of the merge operation is the section in which the page data of the data block is copied. The present invention can reduce the time of the copy interval by selectively copying the page of the data block.

For example, if there are x valid pages in the log block, y pages in which data exists in the data block, and z is the time taken to copy one page, the total time required for the merge operation is (x + y) * z In this case, if the number of deleted pages is i from the viewpoint of the file system, i * z time is unnecessary time in the merge execution time. Therefore, the merge execution time reduced by the user device and the merge method according to the present invention becomes (i * z- (time to read the FAT area)).

Meanwhile, the user device 2000 illustrated in FIG. 6 may perform an invalidation operation by using the above-described logging method. That is, the user device 2000 applies a logging method that records the location of the area to be deleted when the size of the data to be deleted exceeds the reference size. If the data is not exceeded, the user device 2000 may immediately invalidate the data to be deleted. In addition, as described with reference to FIG. 5, the user device 2000 may perform an invalidation operation according to whether to delete general data or secure data.

III. User device invalidating data in buffer memory

9 is a block diagram illustrating an example of a user device for invalidating data in a buffer memory. Referring to FIG. 9, the user device 3000 includes a host 3100 and a storage device 3200. The host 3100 and the storage device 3200 may be connected by a standardized interface such as a USB, SCSI, ESDI, SATA, PCI-express, or IDE interface.

The host 3100 includes a central processing unit (CPU) 3110 and a memory 3120. At least a portion of the memory 3120 includes a main memory of the host 3100. Alternatively, the memory 3120 may be configured as a main memory of the host 3100.

Application program 3121 and file system 3122 are provided in memory 3120, respectively. File system 3122 includes all file systems including a file allocation table (FAT) file system. However, the file system is not limited to this, and it is obvious to those who have acquired the general knowledge in this field.

When all or part of the file data processed by the application program 3121 is deleted, the CPU 3110 may provide an invalidity command to the storage device 3200. Address information and data size information for designating the deleted data together with the invalidation command may be transmitted to the storage device 3200.

The FAT file system includes a master boot record (MBR), partition boot record (PBR), first and second file allocation tables (primary FAT, copy FAT), and root directory. can do. The file to be stored / stored in the storage device 3200 may be identified using two pieces of information. The information is the file name of the file and the path of the directory tree it passes through to reach where the file is stored.

Each entry in a directory is, for example, 32-bytes long and stores information such as file name, extension, file attribute byte, date / time last changed, file size, and connection to the startup cluster. . In particular, in the file name, one special code is used as the first character of the file name to indicate the deleted file.

For example, place the hexadecimal byte code E5h in place of the first letter of the file name. This is a special tag that tells the system that this file has been deleted. When any file is deleted, the CPU 3110 places one special code in the first character of the file name of the deleted file, as well as an invalidation command (or invalidation for deleted file data) corresponding to the deleted file. Information) to the storage device 3200.

9, the storage device 3200 includes a storage medium 3220, a buffer memory 3240, and a controller 3260. The storage device 3200 is a buffer memory when invalid information (for example, including an invalidation command, address information of file data deleted at a higher level of the storage device, and size information of deleted data) is input from the outside. The data of 3220 (which indicates data that has already been deleted at a higher level of the storage device) is configured to be prevented from being written to the storage medium 3260. This will be explained in detail later.

The storage medium 3220 is for storing data (including document data, image data, music data, and data of any format that can be stored, such as a program), and may be configured as a nonvolatile memory such as a magnetic disk or a flash memory. Can be. However, it is apparent to those skilled in the art that storage media 3220 is not limited to those disclosed herein.

Subsequently, buffer memory 3240 is used for smooth data transfer between host 3100 and storage medium 3220 and includes high speed volatile memory such as DRAM or SRAM or MRAM, PRAM, FRAM, NAND flash memory, or NOR. It may be composed of a nonvolatile memory such as a flash memory. For example, the storage medium 3220 and the buffer memory 3240 may be identically configured with a memory such as a flash memory.

The buffer memory 3240 may operate as a write buffer. For example, the buffer memory 3240 may operate as a write buffer for temporarily storing data to be used in the storage medium 3220 at the request of the host 3100. In addition, the function of the write buffer can optionally be used. For example, in some cases, data transferred from the host 3100 may be directly transmitted to the storage medium 3220 without passing through the write buffer, that is, the buffer memory 3240. This function of the storage device 3200 is called a write bypass function.

Alternatively, the buffer memory 3240 may operate as a read buffer. For example, the buffer memory 3240 may operate as a read buffer for temporarily storing data read from the storage medium 3220 at the request of the host 3100. Although only one buffer memory is shown in the figure, two or more buffer memories may be provided. In this case, each buffer memory can be used as a write buffer, a read buffer or a buffer having both functions.

The controller 3260 is configured to control the storage medium 3220 and the buffer memory 3240. When a read command is input from the host 3100, the controller 3260 controls the storage medium 3220 to move data stored in the storage medium 3220 to the host 3100. Alternatively, when a read command is input from the host 3100, the controller 3260 may move the data stored in the storage medium 3220 to the host 3100 through the buffer memory 3240 to the host 3100 and the buffer memory. Control 3240.

When a write command is input from the host 3100, the controller 3260 causes data related to the write command to be temporarily stored in the buffer memory 3240. All or part of the data temporarily stored in the buffer memory 3240 is stored in the controller 3260 when the free space of the buffer memory 3240 is insufficient or when an idle time (an idle time of the controller 3260 occurs when there is no request from the host) occurs. ) Is transferred to the storage medium 3220.

In order to perform the above operation, the controller 3260 indicates whether address mapping information between the storage medium 3220 and the buffer memory 3240 and data stored in the buffer memory 3240 are valid information. It includes a mapping table (3261) for storing the write status information.

The write status information is updated according to invalid information provided from the outside, which will be described later in detail. The controller 3260 controls the storage medium 3220 and the buffer memory 3240 so that all or part of the data stored in the buffer memory 3240 is written to the storage medium 3220 according to the write state information of the mapping table 3331. This will be explained in detail later.

As can be seen from the above description, the storage device 3200 uses the mapping table 3331 when the free space of the buffer memory 3240 is insufficient or when an idle time (an idle time of the controller 3260 occurs when there is no request from the host) occurs. The whole or a part of the data stored in the buffer memory 3240 is not transferred to the storage medium 3220 with reference to the write state information of FIG.

That is, the storage device 3200 receives and inputs invalid information indicating whether the data stored in the buffer memory 3240 is valid data or invalid data from the outside of the storage device 3200 (eg, the host). Reference is made to prevent invalid data among the data stored in the buffer memory 3240 from being used in the storage medium 3220.

In other words, the storage device 3200 selectively tags valid / invalid information on data stored in the buffer memory 3240 to selectively control data write / write operations to the storage medium 3220. Accordingly, the write performance of the storage device 3200 is improved, the lifespan of the storage medium 3220 due to unnecessary write operations can be prevented, and power consumption due to unnecessary write operations can be reduced.

10 and 11 exemplarily illustrate mapping tables of the controller 3260 illustrated in FIG. 9. 10 and 11, reference numeral “BBN” denotes a block number of the buffer memory 3240, reference numeral “DCN” denotes a cluster number of the storage medium 3220, and reference numeral “WSI” denotes a buffer memory 3240. ) Represents write status information indicating whether or not the data stored in the "

For convenience of explanation, assume that the block size of the buffer memory 3240 is equal to the size of a cluster composed of a plurality of sectors. However, it will be apparent to those of ordinary skill in the art that the allocation unit of the storage medium 3220 is not limited to those disclosed herein. For example, it is apparent to those skilled in the art that the allocation unit of the storage medium 3220 may be designated as a sector of a magnetic disk or a page, sector, or block of a flash memory. 10 and 11, invalid data is indicated by 'X', and valid data is indicated by 'V'.

As shown in FIG. 10, assume that data associated with three files FILE1, FILE2, and FILE3 are stored as valid data in the buffer memory 3240. Each data may be data read from the storage medium 3220 and updated by the host 3100 and not yet stored in the storage medium 3220. Alternatively, the data may be data transmitted from the host 3100 and not yet stored in the storage medium 3220.

As described above, the file data FILE1, FILE2, and FILE3 stored in the buffer memory 3240 are stored under the control of the controller 3260 when the free space of the buffer memory 3240 is insufficient or when an idle time occurs. Will be moved to media 3220. The controller 3260 is configured to update write state information of the stored files data FILE1, FILE2, and FILE3 according to invalid information transmitted from the host 3100.

For example, if invalid information is input from the host 3100 indicating that the file data FILE2 has been deleted from the host 3100 as a higher level of the storage device 3200, the controller 3260 may be configured as shown in FIG. 11. The write status information of the file data FILE2 is indicated by 'X' to indicate invalid data. In addition, the controller 3260 displays write state information of data written in the storage medium 3220 as 'X' to indicate invalid data.

FIG. 12 is a flowchart for describing an operation of invalidating buffer memory data of the user device illustrated in FIG. 9. The invalidation operation of the user device 3000 illustrated in FIG. 9 includes step S3100 for determining whether an invalidation command is applied from the outside of the storage device 3200, and a temporary operation in the buffer memory 3240 when the invalidation command is input. In step S3120 to invalidate all or part of the stored data. Thereafter, the invalidated data is not written to the storage medium 3220.

13 to 15 are diagrams for describing an invalidation operation of the user device illustrated in FIG. 9. As described above, the controller 3260 of the storage device 3200 refers to the mapping table 3331 to the buffer memory 3240 in order to secure the free space of the buffer memory 3240 or to utilize the idle time. The stored data files are transferred to the storage medium 3220.

As shown in FIG. 13, assume that three file data FILE1-FILE3 are stored as valid data in the buffer memory 3240. At a specific point in time, the controller 3260 of the storage device 3200 refers to the write status information WSI of the mapping table 3331 associated with the stored file data FILE1-FILE3 to determine whether the data is invalid data. do.

As shown in Fig. 13, since all file data FILE1-FILE3 are represented as valid data in the mapping table 3331, the controller 3260 stores the file data FILE1-FILE3 in a storage medium in a predetermined manner. The storage medium 3220 and the buffer memory 3240 are controlled to be moved to the corresponding space of 3320.

If invalid information (including an invalidation command, address information of an invalid data file, and size information of invalid file data) is input from the outside of the storage device 3200 before a data write / write operation to the storage medium 3220. The controller 3260 of the storage device 3200 invalidates file data related to the invalid information (eg, FILE2).

That is, as shown in FIG. 14, the write state information WSI of the mapping table 3331 associated with the file data FILE2 is updated under the control of the controller 3260 to indicate invalid data. Subsequently, the controller 3260 of the storage device 3200 refers to the write status information WSI of the mapping table 3331 related to the stored file data FILE1-FILE3 at a specific point in time to determine which data files are invalid data. Determine whether or not.

As shown in Fig. 14, since only the file data FILE1 and FILE3 are displayed as valid data in the mapping table 3331, the controller 3260 determines that the file data FILE1 and FILE3 are stored in a storage medium ( The storage medium 3220 and the buffer memory 3240 are controlled to be moved to the corresponding space of 3220. That is, the controller 3260 prevents the file data FILE2 from being moved to the corresponding space of the storage medium 220. The space of the buffer memory 3240 processed as invalid file data may be used for subsequent write / read operations.

As another example, assume that only one file data FILE1 is stored in the buffer memory 3240, as shown in FIG. If an invalidation command is input from the outside of the storage device 3200 before the data write / write operation to the storage medium 3220, the controller 3260 of the storage device 3200 invalidates the data FILE1 related to the input invalidation command. Let's do it.

That is, as shown in FIG. 15, the write state information WSI of the mapping table 3331 associated with the data FILE1 is updated under the control of the controller 3260 to indicate invalid data. Thereafter, the controller 3260 of the storage device 3200 determines whether the stored data is invalid data by referring to the write status information WSI of the mapping table 3331 related to the stored data FILE1 at a specific point in time. do.

As shown in FIG. 15, since there is no data displayed as valid data in the mapping table 3331, no rewrite operation occurs in the storage device 3200. In particular, the write operation of the invalid data in the idle state can be prevented. The space of the buffer memory 3240 processed as invalid data may be used for a subsequent write operation.

Although data marked as invalid data is written to the storage medium 3220, it should be noted that a file associated with invalid data which is data stored in the storage medium 3220 is not affected. In addition, the invalidated data may be selectively transferred to the storage medium 3220 under the control of the controller 3260. For example, even when data of the buffer memory 3240 is invalidated according to an invalidation command, the controller 3260 may be configured to perform a write / write operation of selectively invalidated data.

The storage device 3200 illustrated in FIG. 9 refers to a mapping table including write state information (WSI) indicating whether data stored in the buffer memory 3240 is invalid data, and temporarily stores the buffer in the buffer memory 3240. Controls an operation of transferring the stored data to the storage medium 3220. As mentioned above, the write state information of the data is provided from the outside of the storage device 3200, and the data is data generated and transmitted outside of the storage device 3200.

Alternatively, the data may be new data read from the storage medium 3220 and changed outside of the storage device 3200. The storage device 3200 may be applied not only to a user device but also to devices (eg, MP3 players, portable electronic devices, or the like based on a hard disk or flash memory) for storing data. Is evident to those who have acquired common knowledge in this field.

16 is a block diagram illustrating another embodiment of a user device for invalidating data in a buffer memory. Referring to FIG. 16, the user device 4000 includes a host 4100 and an external storage device 4200.

The host 4100 includes a processing unit 4110, a memory 4120, and a buffer memory 4130. Processing unit 4100 will include a central processing unit (CPU), a microprocessor, and the like. Preferably, the processing unit 4110 will be implemented as a central processing unit. The processing unit 4110 will be configured to control the overall operation of the host 4100.

The host 4100 may be configured to perform the role of the controller 3260 described in FIG. 8. For example, the processing unit 4110 may be configured to prevent data in the buffer memory 4130 from being written to the external storage device 4200 in accordance with the mapping table 4124 of the memory 4120. This will be explained in detail later.

With continued reference to FIG. 16, at least a portion of the memory 4120 includes a main memory of the host 4100. Alternatively, the memory 4120 constitutes a main memory of the host 4100. The memory 4120 is provided with an application program 4121 and a file system 4122, respectively. File system 4122 includes all file systems, including the FAT file system. However, the file system is not limited to this, and it is obvious to those who have acquired the general knowledge in this field.

In the memory 4120, a device driver 4123 and a mapping table 4124 will also be further provided. The device driver 4123 is used for controlling and interface of the external storage device 4200. The processing unit 4110 will control the interface with the external storage device 4200 using the device driver 4123.

In addition, the processing unit 4110 may be configured to manage address mapping between the external storage device 4200 and the buffer memory 4130 using the device driver 4123. The mapping table 4124 stored in the memory 4120 stores interface information with the external storage device 4200, address mapping information between the external storage device 4200 and the buffer memory 4130, and data stored in the buffer memory 4130. It will be used to store write status information indicating whether the information is valid.

The write status information will be updated by the processing unit 4110. For example, when all of the data of the file processed by the application 4121 is deleted or when a part of the file processed by the application 4121 is deleted, the processing unit 4110 is the device driver 4123. On the basis of this, the write status information of the mapping table 4124 is updated.

The processing unit 4110 may use the buffer memory 4130 and the external storage device 4200 to write all or a portion of data stored in the buffer memory 4130 to the external storage device 4200 according to the write state information of the mapping table 4124. Will be controlled. Therefore, it is possible to prevent the data of the buffer memory 4130 corresponding to the data that has already been deleted from being written to the external storage device 4200.

Subsequently, the buffer memory 4130 is used for smooth data transfer between the user device 4100 and the external storage device 4200, and may include high speed volatile memory such as DRAM or SRAM, or MRAM, PRAM, FRAM, NAND flash memory, Or a nonvolatile memory such as NOR flash memory. Preferably, the buffer memory 4130 will be implemented as a NAND flash memory.

The buffer memory 4130 can be used as a write buffer. For example, the buffer memory 4130 may be used as a write buffer for temporarily storing data to be used for the external storage device 4200 at the request of the processing unit 4110. In addition, the function of the write buffer can optionally be used. For example, in some cases, data processed by the processing unit 4110 may be directly transferred to the external storage device 4200 without passing through the write buffer, that is, the buffer memory 4130.

The buffer memory 4130 may be used as a read buffer. For example, the buffer memory 4130 may be used as a read buffer for temporarily storing data read from the external storage device 4200 according to a request of the processing unit 4110. Although only one buffer memory is shown in the figure, two or more buffer memories may be provided. In this case, each buffer memory can be used as a write buffer, a read buffer or a buffer having both functions.

16, the external storage device 4200 is for storing data (including document data, image data, music data, and data in any format that can be stored, such as a program), and a magnetic disk or flash memory. It may be composed of a nonvolatile semiconductor memory such as.

The buffer memory will not be provided to the external storage device 4200. This is because the buffer memory of the host 4100 is used as the cache memory which is a write buffer / read buffer. In this case, the buffer memory 4130 and the external storage device 4200 will function as a hybrid hard disk (HHD).

With continued reference to FIG. 16, the processing unit 4110 is configured to control the external storage device 4200 and the buffer memory 4130. The processing unit 4110 controls the external storage device 4200 using the device driver 4123 to move data stored in the external storage device 4200 to the host 4100 as necessary. Alternatively, the processing unit 4110 may use the device driver 4123 and the external storage device 4200 to move data stored in the external storage device 4200 to the host 4100 through the buffer memory 4130 as needed. The buffer memory 4130 is controlled.

The processing unit 4110 allows data to be stored in the external storage device 4200 to be temporarily stored in the buffer memory 4130. All or part of the data temporarily stored in the buffer memory 4130 may be stored in the external storage device under the control of the processing unit 4110 when the free space of the buffer memory 4130 is insufficient or when the idle time of the processing unit 4110 occurs. Is moved to (4200).

In order to perform the above operation, as mentioned above, the processing unit 4110 may determine whether address mapping information between the external storage device 4200 and the buffer memory 4130 and data stored in the buffer memory 4130 are valid information. A mapping table 4124 for storing write state information indicating whether the data is stored is managed.

The host 4100 and the external storage device 4200 may be connected by a standardized interface such as an ATA, SATA, USB, SCSI, ESDI, PCE-express, or IDE interface. However, the interface method for connecting the host 4100 and the external storage device 4200 is not limited thereto. Those skilled in the art will appreciate.

As can be seen from the above description, the user device 4000 according to an embodiment of the present invention writes the mapping table 4124 when the free space of the buffer memory 4130 is insufficient or when the idle time of the processing unit 4110 occurs. The whole or a part of data stored in the buffer memory 4130 is not transferred to the external storage device 4200 with reference to the state information.

That is, the user device 4000 according to an exemplary embodiment of the present invention may use the buffer memory based on the write state information of the mapping table 4124 indicating whether the data stored in the buffer memory 4130 is valid data or invalid data. The invalid data among the data stored in the 4130 is prevented from being used in the external storage device 4200.

In other words, the user device 4000 according to an embodiment of the present invention selectively controls data rewriting operations to the external storage device 4200 by tagging valid / invalid information to data stored in the buffer memory 4130. . The operation of transferring data stored in the buffer memory 4130 to the external storage device 4200 is the same as described with reference to FIGS. 10 to 15. Therefore, the write performance of the user device 4000 is improved, and the lifespan of the external storage device 4200 due to unnecessary write operations can be prevented from being shortened. In addition, battery life can be extended by reducing power consumption due to unnecessary write operations.

In some embodiments, the buffer memory 4130 may be mounted on the host 4100 in an on-board type. Alternatively, the buffer memory 4130 may be connected to the host 4100 through a PCI bus or a PCI-E bus. However, the method of connecting the buffer memory 4130 is not limited to those disclosed herein, and it will be apparent to those who have acquired a general knowledge in the art. For example, it is apparent to those who have acquired common knowledge in this field that all possible interfaces of desktop and notebook computers can be used.

When the buffer memory 4130 is implemented with a nonvolatile memory such as a NAND flash memory or a NOR flash memory, the buffer memory 4130 may be used in various ways. For example, the buffer memory 4130 may be used as a boot-up memory for storing boot code used for booting. This means that the buffer memory 3240 shown in FIG. 9 may also be used as the boot-up memory. In addition, critical software may be stored in the buffer memory 4130 to improve system performance.

Meanwhile, the user device 3000 or 4000 illustrated in FIG. 9 or 16 may perform the invalidation operation by using the aforementioned logging method. That is, the user device 3000 or 4000 applies a logging method that records the location of the area to be deleted when the size of the data to be deleted stored in the buffer memory exceeds the reference size. You can invalidate it immediately. In addition, as described above with reference to FIG. 5, the user device 3000 or 4000 may perform an invalidation operation according to whether to delete general data or secure data.

Ⅳ. User device performing an invalidation operation by using an invalidation command

17 is a block diagram illustrating an example of a user device performing an invalidation operation using a file delete command. Referring to FIG. 17, the user device 5000 may include a host 5200 and a storage device 5205 connected through an interface 5210.

The interface 5210 may be a standard interface such as ATA, SATA, PATA, USB, SCSI, ESDI, IEEE 1394, IDE, PCI-express and / or card interface. The host 5200 includes a processor 5215 in communication with the memory 5220 via an address / data bus 5225.

By way of example, processor 5215 may be a commercially available or custom processor. The memory 5220 is one or more general memory devices including software and data for operating the user device 5000. The memory 5220 may include devices in the form of a cache, a ROM, a PROM, an EPROM, an EEPROM, a PRAM, a flash memory, an SRAM, and a DRAM.

As shown in FIG. 17, memory 5220 includes five or more software and / or data categories. The software and / or data category may be an operating system (OS) 5228, an application 5230, a file system 5235, a memory manager 5240, and an input / output driver 5245.

Operating system 5228 controls the operation of host 5200. More specifically, operating system 5228 may control software and / or hardware resources of host 5200 and may control program execution by processor 5215. The application 5230 represents various application programs that are executed on the host 5200.

File system 5235 stores or organizes computer files and / or data in a storage area, such as memory 5220 and / or storage 5205. File system 5235 is used according to a particular operating system 5228 running on host 5200. The memory manager 5240 controls a memory access operation performed in the memory 5220 and / or a memory access operation performed in an external device such as the storage device 5205. The input / output driver 5245 performs information transfer between another device, such as the storage device 5205, a computer system, or a network (eg, the Internet) and the host 5200.

According to various embodiments of the present disclosure, the host 5200 may be a personal digital assistance (PDA), a computer, a digital audio player, a digital camera, and a mobile terminal.

The storage device 5205 includes a controller 5250 that communicates with the memory 5255 via the address / data bus 5260. The memory 5255 may be various types of memory in which an erase operation is performed before a write operation. For example, the storage device 5205 according to the present invention may be a memory card device, SSD device, ATA bus device, SATA bus device, multimedia card device, SD device, memory stick device, hard disk drive device, hybrid drive device, or It may be a general purpose serial bus flash device.

Controller 5250 includes a processor 5265 that communicates with local memory 5270 via an address / data bus 5175. By way of example, processor 5265 may be a commercially available or custom microprocessor.

Local memory 5270 may be one or more general memory devices including software and data for operating storage device 5205 according to the present invention. Local memory 5270 may include cache, ROM, PROM, EPROM, EEPROM, PRAM, flash memory, SRAM, and DRAM.

As shown in FIG. 17, local memory 5270 includes three or more software and / or data categories. The software and / or data category may be an operating system 5278, a Flash Transition Layer (FTL) module 5280, and a table 5285. Operating system 5278 controls the general operation of storage 5205. More specifically, operating system 5278 may control software and / or hardware resources of storage 5205 and may control program execution by processor 5265. Meanwhile, according to an exemplary embodiment of the present disclosure, the local memory 5270 may not include an operating system 5278.

The flash translation layer module 5280 may be used in a flash memory device. As described above, the flash chip is erased block by block. A typical flash memory device may perform about 100,000 erase operations per block. To prevent some blocks of flash memory from wearing out faster than others, the flash memory device distributes erase cycles throughout the memory. This is called wear management. For wear management, the flash translation layer module 5280 interfaces between the file system 5235 and the storage location of the file / data in the memory 5255. Thus, the file system 5235 does not need to keep track of the actual storage location of the file / data in the memory 5255.

Table 5285 may be maintained by flash translation layer module 5280 and may be used to associate a physical address of a memory allocation unit of memory 5255 with an indicator that indicates whether the memory allocation unit contains invalid data.

18 is a diagram illustrating a data structure in accordance with the present invention that associates memory allocation units of storage 5205 with an indicator indicating whether memory allocation units contain valid or invalid data. In FIG. 18, the memory allocation unit is a page, and the block is shown to include four pages.

Referring to FIGS. 17 and 18, table 5285 associates the physical addresses of pages of flash memory 5255 with logical addresses used by file system 5235. In addition, the table 5285 includes columns indicating whether a particular page of the flash memory 5255 includes invalid data or valid data.

In the embodiment illustrated in FIG. 18, a block of pages including logical addresses 0-3 includes invalid data and thus may be erased. Table 5285 may be used to start an erase operation when all pages in a block contain invalid data.

In the conventional example, when a second write operation is attempted to the logical page address 0, it may be determined whether the logical page address 0 contains invalid data. However, it is not clear whether logical page addresses 1-3 also contain invalid data. Therefore, in order to perform a write operation to the logical page address 0, the entire block including the logical page addresses 0-3 must be erased, so that the data in the logical page addresses 1-3 can be moved to another storage location. Must be copied. If the logical page address 1-3 contains invalid data, such a copy operation may not be required.

To reduce unnecessary copying operations as described above, table 5285 may provide an indicator that indicates pages containing invalid data. In FIG. 18, the data structure according to the present invention is shown in the form of a table, but it will be understood that the data structure according to the present invention can be variously implemented in other forms of data structure.

Computer program code for performing operations of the user device described with reference to FIG. 17 may be written in a high level programming language such as JAVA, C, and / or C ++. In addition, the computer program code for performing an operation according to an embodiment of the present invention may be written in an interpreted language. In order to improve operating performance and / or memory usage, some modules or routines may be written in assembly language or micro-code. The functionality of some or all of the program modules may be implemented as individual hardware components, one or more application specific integrated circuits (ASICs), or programmed digital signal processors or microcontrollers.

In the following, with reference to the flow, flowchart and / or block diagram of a message showing a method, system, apparatus and / or computer program product according to an embodiment of the invention, the invention is described. The flow, flow chart and / or block diagram of the messages illustrate common operations for operating a data processing system including an external data storage device. Each of the figures in the flow, flow and / or block diagrams of messages and combinations thereof may be implemented by computer program instructions and / or hardware operations. The computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus. Computer program instructions may be performed through a computer or programmable data processing apparatus to provide a means for implementing the functions described in the flow, flow, and / or block diagram of a message.

Computer program instructions may be stored in a memory usable or computer readable, and may cause the computer or other programmable data processing instruments to operate in a particular manner. That is, instructions stored on a computer usable or computer readable memory may provide instructions that implement the functionality described in the flow, flow, and / or block diagram of a message.

Computer program instructions may be loaded into a computer or other programmable data processing apparatus to provide a process to be performed by the computer, inducing a series of operational steps to be performed on the computer or other programmable instrument. In other words, instructions executed on a computer or other programmable instrument provide the operational steps for implementing the functions described in the flow, flow and / or block diagram of a message.

19 to 23 are flowcharts for describing operations of the user device illustrated in FIG. 17. Referring to FIG. 19, in operation S5400, the host 5200 transmits an invalidity command for one or more files to the external storage device 5205. The external storage device includes a memory device in which an erase operation is performed before a write operation, such as a flash memory device.

According to the embodiment of the present invention illustrated in FIG. 20, in operation S5500, an invalidation operation is detected at the host 5200. For example, the detection operation may be performed by detecting that metadata associated with the file system is updated with a delete code for the deleted file. When the invalidation operation is detected at the host 5200, in operation S5505, the invalidation command is transmitted to the external storage device 5205. The invalidation command may indicate the logical address associated with the deleted file and the data to be invalidated.

Referring to FIG. 21, in operation S5600, when an invalidation command is received from the host 5200, an invalidation operation is started in the external storage device 5205. The invalidation command may indicate logical addresses and invalidated data of one or more files. In step S5605, according to the logical address and the data to be invalidated, the storage device 5205 identifies whether one or more memory allocation units of the memory 5255 contain invalid data.

Referring to FIG. 22, in step S5700, the flash translation layer module 5280 maintains a data structure such as the table 5285 shown in FIG. 18. The data structure associates logical and physical addresses of memory allocation units of memory 5255. In addition, the data structure may include an indicator indicating whether various memory allocation units contain invalid data. In step S5705, if the physical address of the memory allocation unit is identified as being associated with the invalidated file, the flash translation layer module 5280 updates the data structure to indicate that the identified memory allocation unit contains invalid data. do.

Since various memory operations are performed on the storage device, a garbage collection operation is required to further free up available and adjacent memory blocks. According to an embodiment of the present disclosure, the flash translation layer module 5280 may operate until the operating system 5228 of the host 5200 or the operating system 5280 of the storage device 5205 performs a periodic garbage collection operation. Without waiting, table 5285 can be used to determine the timing to collect memory units containing invalid data.

Referring to FIG. 23, in step S5800, the flash translation layer module 5280 determines whether each of the read / write operation units is an invalid data field, and erases an operation unit (eg, a block of a flash memory device). It is determined whether all read / write operation units (for example, pages of a flash memory device) in the table contain invalid data.

In operation S5805, when it is determined that all read / write operation units in the erase operation unit include invalid data, the erase operation of the erase operation unit may be performed. In this manner, when the erase operation unit is ready to erase, the actual physical file data may be erased. That is, performing physical erase on a file in a storage device in accordance with the present invention is performed faster than waiting for the physical erase to be performed due to a duplicate file write operation performed on the storage device. Thus, it can be applied to applications containing personal or sensitive data.

The flowcharts of FIGS. 19 through 23 illustrate embodiments of the method, system, and structure, function, and operation of a computer program product for operating a user device including an external storage device. In this regard, each block represents a module, segment, or part of code. A module, segment, or portion of code includes one or more executable instructions for implementing a particular logical function. The order of the functions performed in the blocks may differ from that shown in FIGS. 19 to 23. For example, two blocks shown in succession may be performed simultaneously or in reverse order, depending on the function.

Meanwhile, the user device 5000 illustrated in FIG. 17 may perform an invalidation operation by using the aforementioned logging method. That is, the user device 5000 applies a logging method that records the location of the area to be deleted when the size of the data to be deleted exceeds the reference size. If the data is not exceeded, the user device 5000 may immediately invalidate the data to be deleted. In addition, as described with reference to FIG. 5, the user device 5000 may perform an invalidation operation according to whether to delete general data or secure data.

Ⅴ. User Device Performing Self Invalidation Behavior

24 is a block diagram illustrating a storage device that performs self invalidation operation without host intervention. In FIG. 24, as an example of a storage device, a solid state drive (SSD) will be described. Referring to FIG. 24, the SSD 6000 includes an SSD controller 6200 and a nonvolatile storage medium 6400. The SSD controller 6200 includes first and second interfaces 6210 and 6230, a controller 6220, and a memory 6240.

The first interface 6210 functions as a data input / output interface with the host. The first interface 6210 is a non-limiting example, and includes a universal serial bus (USB) interface, an ATA interface, a SATA interface, a SCSI interface, and a PCI-express.

The second interface 6230 performs a data input / output interface function with the nonvolatile storage medium 6400. In particular, second interface 6230 is used to send / receive various commands, addresses and data to / from nonvolatile storage medium 6400. A variety of different structures and configurations of the second interface 6230 are possible, and detailed description thereof is omitted here for brevity.

The controller 6220 and the memory 6240 are connected between the first and second interfaces 6210 and 6230, and both control the flow of data between the host (not shown) and the nonvolatile storage medium 6400. Has a function to manage The memory 6240 includes, for example, DRAM or SRAM. The controller 6220 includes, for example, a central processing unit (CPU), a direct memory access (DMA) controller, and an error correction control (ECC) engine.

24, the nonvolatile storage medium 6400 of this embodiment may include a high speed nonvolatile memory 6410 and a low speed nonvolatile memory 6620. However, the embodiment described above is not limited to structures having dual-speed memories. That is, the nonvolatile storage medium 6400 may be composed of a single type of memory that operates at a single speed. The high speed nonvolatile memory 6410 can operate at a relatively high speed (eg, an arbitrary write speed) when compared to the low speed nonvolatile memory 6620.

In an embodiment, the high speed nonvolatile memory 6410 may be a single-level cell (SLC) flash memory, and the low speed nonvolatile memory 6620 may be a multi-level cell (MLC) flash memory. However, the present invention is not limited to this. For example, the fast nonvolatile memory 6410 may be a phase change random access memory (PRAM) or an MLC flash memory that stores one bit per cell. The high speed nonvolatile memory 6410 and the low speed nonvolatile memory 6420 may be configured of the same type of memory. Here, the operation speed is different by fine-grain mapping to the high speed nonvolatile memory 6410 and coarse-grain mapping to the low speed nonvolatile memory 6620.

In general, high speed nonvolatile memory 6410 is used to store frequently accessed (written) data, such as metadata, and low speed nonvolatile memory 6420 stores relatively less accessed (written) data, such as media data. Used to store

That is, the data write frequency in the high speed nonvolatile memory 6410 is statistically higher than the data write frequency in the low speed nonvolatile memory 6620. Also, depending on the nature of each data stored, the storage capacity of low speed nonvolatile memory 6620 is typically higher than that of high speed nonvolatile memory 6410. In other words, however, the embodiments described herein are not limited to the use of two or more memories operating at different speeds.

25 illustrates an example of logically partitioning a storage area of the nonvolatile storage medium 6400. Referring to FIG. 25, the first sector of the solid state drive includes a master boot record (MBR), and the remaining sectors are divided into a plurality of partitions. In addition, each of the partitions typically contains a boot record at the end of the logic.

FIG. 26 shows a well-known 512 byte example of the MBR shown in FIG. 25. In general, the MBR is used to maintain, for example, the primary partition table of a solid state drive (SSD). It is also used in boot strapping operations to perform code instructions present in the MBR after a bias transmission execution of the user device. MBR is also used to uniquely recognize individual storage media.

FIG. 27 shows an example of the layout of a single 16 byte partition record of the MBR shown in FIG. 26. In the example of the IBM partition table standard, four of the partition records shown in FIG. 27 include the partition table of the MBR.

28 is a table showing partition types and corresponding ID values. In this regard, the operating system (OS) can create multiple partitions from further detailed primary partitions. These partitions are referred to as "extended partitions". Each partition created on an extended partition is called a logical partition, and each logical partition remodels the same or different file systems.

The MBR technology described above is one of several standards in an increasingly developing industry. For example, the Extensible Firmware Interface (EFI) standard is an alternative technology for the PC bias standard. PC bias, on the other hand, uses the MBR technique described above, and the EFI standard uses GUID Partition Table (GPT) as the standard for the layout of partition tables in logically partitioned solid state drives. The invention is not limited to any particular partition standard.

Data included in the MBR (or GUID) partition table of FIG. 26 is an example of "storage-level" metadata. For example, metadata associated with logical storage areas of a solid state drive (SSD). This is in contrast to “file system level” metadata, which is metadata associated with the file system of the user device (or computer system). Examples of file systems include file allocation tables (FAT), new technology file systems (NTFS), second and third extended file systems (ext2 and ext3).

That is, when a user deletes a file from nonvolatile storage medium 6400, a file system operating on the system processes the invalidation command. At this point, the file of the nonvolatile storage medium 6400 is deleted from the user's point of view. In practice, however, the file system does not change the file data in physical memory, but instead considers it "invalid". The host contains an application for communicating with the file system. The flash translation layer (FTL) tracks and maintains the physical locations of the memory units of nonvolatile storage medium 6400 associated with each of the files. Thus, the file system only needs to reference logical memory units.

Embodiments of the invention, described in greater detail below, are controlled to at least partially monitor updated metadata to retrieve locations of invalid data stored in a solid state drive (SSD).

The monitored metadata may be storage level metadata or file system level metadata. For example, in the case of storage level metadata, metadata will be included in the partition table, and invalid data is retrieved according to changes in the metadata of the partition table.

In one embodiment, it is detected whether the partition metadata of the solid state drive (SSD) has changed, and if so, the partition metadata is analyzed to retrieve the location of invalid data stored in the solid state drive (SSD). . This analysis includes determining if the file system type of the partition has changed and invalidating the data corresponding to the changed file system type. Alternatively, or in addition, the analysis may include determining if the partition has changed and invalidating the data in response to the changed partition.

29 and 30 are described with reference to a method of invalidating an erased data area of a solid state drive (SSD) according to an exemplary embodiment of the present invention.

In general, this embodiment relates to the monitoring of metadata contained in a partition table, such as a standard table of primary partitions of an MBR in a BIOS system. In steps S6510 and S6520 of FIG. 29, the MBR address area is monitored to determine whether the MBR address has been accessed. Examples of MBR, primary partitions, and partition record are shown in FIG.

If it is determined that the MBR address is accessed, it is searched whether the partition table has been changed in step S6530. For example, the partition table may change while the partition is partitioned. In this case, all data of the divided partition is invalidated.

In the case of Yes determination in step S6530, the starting position of the partition and the type of file system (partition type) are configured in step S6540. Then, in step S6550, the metadata is analyzed according to the file system type, and the deleted data area is invalidated.

A method of invalidating an erased data area of a solid state drive (SSD) according to an exemplary embodiment of the present invention will be described with reference to FIGS. 31 and 32.

In general, this embodiment relates to the monitoring of metadata contained in a file allocation table (FAT). In particular, by checking the cluster association (or lack thereof), it is determined whether the data associated with the cluster is deleted data.

In general, the file system to be used for storing files in a flash solid state drive (SSD) has a memory allocation unit defined to account for the smallest logical amount of disk space that can be allocated to hold a file. For example, an MS-DOS file system known as a file allocation table (FAT) calls such a memory allocation unit a cluster.

In the method of FIG. 31, the file entry is initially checked in steps S6610 and S6620, and it is determined whether the file entry is [00 00 00 00]. If the determination in step S6620 is Yes, the clusters matched in step S6630 are not connected, and its data is invalidated.

33 and 34 are described with reference to a method of invalidating an erased data area of a solid state drive (SSD) according to an embodiment of the present invention.

In general, this embodiment relates to the monitoring of metadata contained in the New Technology File System (NTFS). Initially, in step S6710, the start of the master file table (hereinafter MFT) from the NTFS boot record is checked. In this example, the bitmap, which is the sixth entry of the MFT, is retrieved in step S6720, and then the bitmap table is checked in step S6730. In step S6740, it is determined whether there is deleted data in the bitmap table, and if the answer is yes, the matched data area is invalidated.

By invalidating the data and data regions as described above, a merge operation is possible in the solid state drive SSD without copying the invalid data. In addition, garbage collection operations can be made more efficient, for example.

35 is a block diagram of a user device 6800 according to an embodiment of the present invention. As shown, the user device 6800 is connected to the bus system 6810 and the read only memory (ROM) that stores the software (eg, BIOS) that is connected to the bus system 6810 and used to initialize the user device 6800. It includes.

User device 6800 also includes random access memory 6830, a central processing unit 6840, and a solid state drive (SSD) system 6850, which all function as working memory, all of which are bus systems 6810. Is connected to. Solid state drive (SSD) system 6850 includes a solid state drive (SSD) and a controller (see, eg, FIG. 24). In addition, in the example shown in FIG. 35, a solid state drive (SSD) system 6850 includes a master boot record and is logically divided into a plurality of partitions.

Meanwhile, the storage device 6000 illustrated in FIG. 24 may perform an invalidation operation by using the above-described logging method. That is, the storage device 6000 applies a logging method that records the location of the area to be deleted when the size of the data to be deleted exceeds the reference size. If the size of the data to be deleted does not exceed the reference size, the storage device 6000 may immediately invalidate the data to be deleted. In addition, as described with reference to FIG. 5, the storage device 6000 may perform an invalidation operation according to whether to delete general data or secure data.

VI. Recovery  User device performing the action

In the case of a hard disk directly controlled by the file system, the deleted data can be recovered by a recovery command of the file system except when overwritten. However, in the case of a flash memory device, an erase operation is performed through both a file system and a flash conversion layer. Accordingly, even when data is recoverable at the file system level, data recovery may not be possible at the flash translation layer level. As a result, a data recovery method in a semiconductor memory device using a conversion layer such as a flash memory device may be problematic.

36 is a block diagram showing a software structure of a user device performing a recovery operation. Referring to FIG. 36, a software structure 7100 of a user device has a software hierarchy in order of an application 7110, a file system 7120, a flash translation layer 7130, and a flash memory 7140.

File system 7120 receives commands from application 7110. Instructions from the application 7110 include a data store command, a data read command, a data move command, a data delete command, and a data recovery command. File system 7120 communicates the logical address of the data to be processed to flash translation layer 7130.

The flash translation layer 7130 receives a logical address from the file system 7120. The flash translation layer 7130 converts a logical address into a physical address. The flash translation layer 7130 references an address mapping table for address translation. The address mapping table includes a plurality of mapping data. Each mapping data defines a correspondence between logical addresses and physical addresses. The physical address is provided to the flash memory 7140.

The flash memory 7110 is divided into a meta data area and a user data area. User data such as text, video, and audio are stored in the user data area. Information regarding the user data is stored in the meta data area. For example, location information of user data is stored in the metadata area. Accordingly, the file system 7120 may determine the storage location of the user data by referring to the metadata area.

In an embodiment according to the present disclosure, the flash translation layer 7130 includes a queue 7132. A queue is a type of buffer that processes data in a first-in first-out (FIFO) manner. That is, the data input first is output first. The flash translation layer 7130 has a separate queue 7122 to prevent data from being lost due to a merge or erase operation. This will be described in detail with reference to the drawings to be described later.

37 is a block diagram illustrating a hardware structure of a user device performing a recovery operation. Referring to FIG. 37, the user device 7200 may include a host 7280, a memory controller 7270, and flash memory devices 7210 to 7260. The flash memory devices 7210 to 7260 may include a memory cell array 7210, a page buffer 7220, a row select circuit 7230, a column select circuit 7240, a data input / output circuit 7250, and a control logic 7260. Include.

The memory cell array 7210 is composed of a plurality of memory cells. The memory cell array 7210 is divided into a user data area 7141 and a meta data area 7212. In the user data area 7141, user data such as text, audio, and video are stored. The meta data area 7212 stores meta data about user data. For example, the metadata stores location information of user data. Although not shown in the drawings, the memory cell array 7210 may be composed of memory cells arranged in a matrix form of rows (or word lines) and columns (or bit lines). The memory cells may be arranged to have a NAND structure or to have a NOR structure.

The page buffer circuit 7220 operates as a sense amplifier or as a write driver. In a read operation, the page buffer circuit 7220 reads data from the memory cell array 7210. In the program operation, the page buffer circuit 7220 drives the bit lines to a power supply voltage or a ground voltage according to data input through the column select circuit 7240.

The row select circuit 7230 is connected to the memory cell array 7210 through a word line. Row select circuit 7230 is controlled by control logic 7260. The row select circuit 7230 drives the selected and unselected rows to the corresponding word line voltages, respectively, in response to the row address. For example, in a program operation, the row select circuit 7230 drives the selected row to the program voltage Vpgm and the unselected rows to the pass voltage Vpass, respectively. In the read operation, the row select circuit 7230 drives the selected row to the read voltage Vread and the unselected rows to the pass voltage Vpass, respectively.

The column select circuit 7240 reads data latched in the page buffer circuit 7220 in response to a column address applied from a column address generation circuit (not shown in the figure), or the page buffer circuit 7220 is used. Pass data to

The data input / output buffer 7250 transfers the data input from the memory controller 7270 to the column select circuit 7240. Alternatively, the data input / output buffer 7250 transfers data input from the column select circuit 7240 to the memory controller 7270.

Control logic 7260 is configured to control the overall operation of the flash memory device. For example, the control logic 7260 is configured to control a program, a read operation, an erase operation, and the like of the flash memory device.

The memory controller 7270 includes a mapping table 7727 and a queue 7726. The flash translation layer 7130 illustrated in FIG. 35 is performed in the form of firmware on the memory controller 7270. The mapping table 7719 stores correspondence relationships between logical addresses and physical addresses. The mapping table 7721 is used by the flash translation layer 7130 to convert a logical address input from the host 7280 into a physical address. The flash translation layer 7130 transfers the physical address to the flash memory device.

The queue 7726 is a type of buffer that stores data in a first-in first-out (FIFO) manner. That is, the data first input to the queue 7122 is output first. The size of the queue 7122 can be changed as needed. In the present invention, the queue 7122 is provided to delay invalidation of the mapping data and to prevent the data from being lost by the merge and erase operations.

The user device 7200 according to an embodiment of the present invention may recover deleted data. Hereinafter, a data recovery method according to the present invention will be described in detail with reference to the drawings.

38 is a flowchart illustrating a data deletion operation. Referring to FIG. 38, the data deletion operation is divided into three steps S7110 to S7130.

In operation S7110, an invalidity command is input from the application 7110. Application 7110 is run on host 7280 of FIG.

In operation S7120, the file system 7120 deletes meta data of data to be deleted in response to an invalidation command. File system 7120 is performed on host 7280 of FIG. 37. File system 7120 may be included in an operating system.

In operation S7130, the flash translation layer 7130 invalidates the mapping table corresponding to the data to be deleted. Hereinafter, a method of deleting metadata and a method of invalidating mapping data corresponding to data to be deleted will be described in detail with reference to the accompanying drawings.

39 is a conceptual view illustrating a method of deleting meta data when data is deleted. 39 (a) shows before the metadata is deleted, and FIG. 39 (b) shows after the metadata is deleted.

Referring to Fig. 39A, user data named "Movie.avi" and "Music.mp3" are stored in the user data area. Meta data corresponding to user data is stored in the meta data area. Each metadata will contain information about the storage location of the corresponding user data. Accordingly, the file system 7120 may access the user data by referring to the metadata.

When a command is input from the application 7110 to delete user data named "Movie.avi", the file system 7120 deletes only metadata corresponding to the user data. File system 7120 does not delete the user data. Referring to FIG. 39B, it can be seen that only metadata is deleted. Accordingly, the file system 7120 recognizes that user data named "Movie.avi" has been deleted. That is, even after deletion, the user data exists in the user data area. In this case, only the correspondence between the meta data and the user data is destroyed. Therefore, when data restoration is required, user data can be accessed normally when restoring deleted metadata.

40 is a block diagram illustrating a method of invalidating mapping data corresponding to data to be deleted during a data deletion operation. 40 (a) shows a mapping table before data deletion, and FIG. 40 (b) shows a mapping table after data deletion.

In an embodiment, it is assumed that physical sectors 1 to 10 (PSN 1 to PSN 10) correspond to data to be deleted. In other words, data to be deleted is stored in physical sector 1 to physical sector 10 (PSN 1 to PSN 10). Referring to FIG. 40A, the flash translation layer 7130 maps a logical sector number to a physical sector number with reference to the mapping table 7171. For example, logical sector 1 (LSN 1) corresponds to physical sector 1 (PSN 1). In addition, the mapping table stores whether a physical sector is valid. For example, physical sector 1 to physical sector 10 (PSN 1 to PSN 10) are designated as valid.

When a data invalidation command is input from the file system 7120, the flash translation layer 7130 invalidates the mapping data corresponding to the data to be deleted. Referring to FIG. 40 (b), it can be seen that the physical sectors 1 to 10 (PSN 1 to PSN 10) are invalid. Accordingly, the flash translation layer 7130 may not convert the logical sectors 1 through 10 (LSN 1 through LSN 10) into physical sectors 1 through 10 (PSN 1 through PSN 10). This also means that the flash translation layer can allocate physical sector 1 to physical sector 10 for storage of other data.

Since only the physical sector is invalidated by the erase operation, data stored in the physical sector is not actually erased. As a result, data exists even after deletion. Thus, when data recovery is required, validating the invalidated physical sectors can normally recover the data.

As described above, when data is deleted, the metadata is deleted by the file system 7120 and the mapping data is invalidated by the flash translation layer 7130. As a result, the delete operation is performed by deleting the metadata and invalidating the mapping data instead of actually deleting the user data. Therefore, it is possible to recover the user data by restoring the meta data and validating the mapping data.

However, as the number of invalid physical sectors increases, the capacity of the flash memory device decreases. Accordingly, the flash memory device collects internally valid physical sectors in another physical sector and erases invalid physical sectors to increase storage capacity. This is called a merge operation. In addition, data stored in the physical sector may be lost by a deletion command from the outside.

41 is a conceptual diagram for explaining a merge operation. FIG. 41A shows the mapping table before merge and FIG. 41B shows the mapping table after merge. Referring to FIG. 41A, it is assumed that the first and third physical sectors of the first block Block 1 and the seventh and eighth physical sectors of the second block Block 2 are invalidated by the erase operation. . Invalidated blocks are subject to merge operation.

Referring to FIG. 41 (b), only the effective sectors of the first block Block 1 and the second block Block 2 are stored in the third block Block 3 by the merge operation, and the first block Block 1 is closed. And the second block Block 2 is erased. When the physical sector is erased by the merge operation, existing data is permanently lost. Erased blocks may be allocated for storage of other data by the flash translation layer 7130.

However, the merge operation is performed regardless of the file system 7120. In order to improve the performance of the system, the merge operation may be performed during a background time from the file system 7120. Thus, the invalidated sectors are at risk of being erased by the merge operation at any time. Therefore, in order to recover data stored in the physical sector, the performing of the merge operation on the physical sector should be suppressed.

The user device 7200 according to an embodiment of the present invention delays invalidation of the mapping data. Alternatively, the user device 7200 prevents invalid physical sectors from being merged and erased. Thus, data stored in the physical sector can be recovered.

In the present invention, an invalid delay queue is used to delay invalidation of mapping data. Physical sectors to be invalidated are sequentially registered in the invalid delay queue. When the invalid delay queue is full, the first registered physical sector is deregistered according to the first in, first out (FIFO) method. Deregistered physical sectors are invalidated. As described above, the data stored in the physical sector can be recovered by delaying invalidation of the mapping data. The size of the invalid delay queue can be changed. For example, if the size of the invalid delay queue is large, invalidation of the mapping data will be delayed for a long time. Conversely, if the size of the invalid delay queue is small, invalidation of the mapping data will be delayed for a short time.

42 is a conceptual diagram for explaining a method of operating an invalid delay queue. Referring to FIG. 42, it is assumed that physical sectors 1, 3, 7, and 8 (PSNs 1, 3, 7, and 8) are invalidated in order by the data erasing operation. In an embodiment according to the present invention, physical sectors 1, 3, 7, and 8 (PSNs 1, 3, 7, and 8) are not immediately invalidated by a delete operation but are registered in the invalid delay queue. In detail, first, physical sector 1 (PSN 1) is registered in the invalid delay queue. At this time, the data stored in the physical sector 1 (PSN 1) is not stored in the invalid delay queue, but only the position of the physical sector 1 is stored. In the same manner, physical sectors 3, 7, 8 (PSN 3, 7, 8) are sequentially registered in the invalid delay queue.

The flash translation layer 7130 does not invalidate the physical sectors registered in the invalid delay queue. Therefore, physical sectors 1, 3, 7, 8 (PSN 1, 3, 7, 8) remain valid. If the invalid delay queue is full, the registration of the first-registered physical sector PSN 1 is released according to the first-in first-out (FIFO) method. Physical sectors deregistered from the invalid delay queue are invalidated. The invalidated physical sectors are subject to merge and erase operations.

FIG. 43 is a flowchart illustrating a data recovery method using the invalid delay queue of FIG. 42. Referring to FIG. 43, a data recovery method according to the present invention is divided into four steps S7210 to S7240.

In operation S7210, the application 7110 may apply a data recovery command to the file system 7120. File system 7120 communicates a data recovery command to flash translation layer 7130. In operation S7220, the flash translation layer 7130 determines whether mapping data corresponding to data to be restored to the invalid delay queue is registered. If the mapping data corresponding to the data to be restored to the invalid delay queue is registered, the S7230 trick is executed, and if not, the data recovery ends. In operation S7230, the flash translation layer 7130 deregisters mapping data corresponding to data to be recovered from the invalid delay queue. In operation S7240, the file system 7120 recovers metadata of data to be recovered.

By restoring the metadata corresponding to the deleted data through the above-described method, stable recovery of the data is possible. The order of recovery of the metadata and update of the invalid delay queue may be changed. For example, the invalid delay queue may be updated after the metadata is recovered first.

In the above embodiment, invalidation of the mapping data is delayed by using an invalid delay queue. However, in the case of a physical sector that is already invalidated, data stored in the physical sector can be protected by preventing the merge and erase operations from being performed.

Another embodiment of the present invention uses a merge / erase protection queue to prevent merge and erase operations on invalidated physical sectors. Invalid physical sectors registered in the merge / erase prevention queue are not merged and erased. Thus, data stored in the physical sector can be maintained. The size of the merge / erase prevention queue can be changed. For example, a large merge / erase prevention queue will prevent merge and erase operations of invalidated physical sectors for a long time. Conversely, if the size of the merge / erase prevention queue is small, the merge and erase operations of the invalidated physical sectors will be prevented for a short time.

44 is a conceptual diagram illustrating a method of operating a merge / erase prevention queue. Referring to FIG. 44, it is assumed that physical sectors 1, 3, 7, and 8 (PSNs 1, 3, 7, and 8) are invalidated in order by a data erasing operation. In an embodiment according to the present invention, physical sectors 1, 3, 7, and 8 (PSNs 1, 3, 7, and 8) are sequentially registered in the merge / erase prevention queue after being invalidated by the erase operation. In detail, the invalidated physical sector 1 (PSN 1) is first registered in the merge / erase prevention queue. At this time, the data stored in the physical sector 1 (PSN 1) is not stored in the merge / erase prevention queue, but only the position of the physical sector 1 (PSN 1) is stored. In the same way, physical sectors 3, 7, 8 (PSN 3, 7, 8) are sequentially registered in the merge / erase prevention queue.

The flash translation layer 7130 excludes the physical sectors registered in the merge / erase prevention queue from the object of the merge and erase operations. Thus, data stored in physical sectors 1, 3, 7, 8 are not merged and erased. If the merge / erase prevention queue is full, the first registration of the physical sector registered according to the first in, first out (FIFO) method is released. Physical sectors deregistered from the invalid delay queue are subject to merge and erase operations.

45 is a flowchart illustrating a data recovery method using the merge / erase prevention queue of FIG. 44. Referring to FIG. 45, the data recovery method according to the present invention is divided into five steps.

In operation S7310, a data recovery command is input from the application 7110 to the file system 7120. File system 7120 communicates a data recovery command to flash translation layer 7130. In operation S7320, the flash translation layer 7130 determines whether mapping data corresponding to data to be restored to the merge / erase prevention queue is registered. If mapping data corresponding to data to be restored to the merge / erase prevention queue is registered, step S7330 is performed. If not, data recovery is terminated.

In operation S7330, the flash translation layer 7130 validates mapping data corresponding to the data to be recovered through the mapping table. In operation S7340, the flash translation layer deregisters mapping data corresponding to data to be recovered from the merge / erase prevention queue. In operation S7350, the file system 7120 recovers deleted metadata of the data to be recovered. By preventing the merge and erase operations of the invalidated physical sectors through the above-described method, data loss is prevented. As a result, reliable recovery of data is possible.

In another embodiment of the present invention, the invalid delay queue and the merge / erase prevention queue are used together. First, the invalidation of the mapping data is delayed through the invalid delay queue. The invalid physical sectors are then registered in the merge / erase prevention queue to prevent merge and erase operations on the invalid physical sectors.

46 is a conceptual diagram for explaining an operation method when an invalid delay queue and a merge / erase prevention queue are used at the same time. 46 (a) shows a case where only an invalid delay queue is used, and FIG. 46 (b) shows a case where an invalid delay queue and a merge / erase prevention queue are used together.

In an embodiment of the present invention, the mapping data corresponding to the data to be deleted is delayed by being registered in the invalid delay queue. When the invalid delay queue becomes full, the registration of the first registered physical sector is released according to the first in, first out (FIFO) method. The mapping data deregistered from the invalid delay queue is registered in the merge / erase prevention queue. The mapping data registered in the merge / erase prevention queue is not the target of merge / erase. Therefore, data stored in the physical sector is not lost by merge and erase for a certain time.

First, referring to FIG. 46 (a), it is assumed that physical sectors 1, 2, 4, and 5 (PSN 1, 2, 4, 5) are sequentially invalidated. In an embodiment according to the present invention, physical sectors are registered in the invalid delay queue before being invalidated. Physical sectors registered in the invalid delay queue are not invalidated. At this time, the merge / erase prevention queue is not used.

46 (b) shows a case in which physical sector 7 (PSN 7) is registered in the invalid delay queue. When the invalid delay queue becomes full, registration of the first registered physical sector 1 (PSN 1) is released according to the first-in first-out method. Physical sector 1 (PSN 1) deregistered from the invalid delay queue is invalidated. The invalidated physical sector 1 (PSN 1) is registered in the merge / erase prevention queue. Physical sectors registered in the merge / erase prevention queue are excluded from the merge and erase operations.

FIG. 47 is a flowchart illustrating a data recovery method using both an invalid delay queue and a merge / erase prevention queue of FIG. 46. Referring to FIG. 47, the data recovery method according to the present invention is divided into seven steps.

In operation S7410, a data recovery command is input from the application 7110 to the file system 7120. File system 7120 forwards the data recovery command to flash translation layer 7130. In operation S7420, the flash translation layer 7130 determines whether mapping data corresponding to data to be restored to the invalid delay queue is registered. If mapping data corresponding to data to be restored to the invalid delay queue is registered, step S7430 is performed, and if not, step S7440 is performed.

In operation S7430, the flash translation layer 7130 deregisters mapping data corresponding to data to be restored from the invalid delay queue. In operation S7440, it is determined whether mapping data corresponding to data to be restored is registered in the merge / erase prevention queue. If mapping data corresponding to data to be restored to the merge / erase prevention queue is registered, step S7450 is performed. If not, data recovery is terminated.

In operation S7450, the flash translation layer 7130 validates the invalidated physical sector through the mapping table. In operation S7460, the flash translation layer 7130 deregisters mapping data corresponding to data to be recovered from the merge / erase prevention queue. In operation S7470, the file system 7120 recovers deleted metadata of the data to be recovered.

Through the above method, data is prevented from being lost by delaying invalidation of the block corresponding to the deleted data and preventing merge and erase operations of the physical sector corresponding to the invalidated data. As a result, stable recovery of data becomes possible.

48 is a block diagram schematically illustrating a user device 7300 including a semiconductor memory device according to an embodiment of the present invention. Referring to FIG. 48, a user device 7300 according to the present invention may include a central memory electrically connected to a semiconductor memory device 7310 and a system bus 7350 including a nonvolatile memory device 7311 and a memory controller 7312. Device 7330, user interface 7340, and power supply 7320.

In the nonvolatile memory device 7311, data provided through the user interface 7340 or processed by the CPU 7330 may be stored through the memory controller 7312. The semiconductor memory device 7310 may be configured as a solid state drive (SSD), and the solid state drive supports recovery of deleted data. In addition, solid-state drive (SSD) products, which are expected to replace hard disk drives (HDDs), are in the spotlight in the next-generation memory market. SSDs have the advantage of being faster, more resistant to external shocks, and lower power consumption than mechanically moving hard disk drives.

Although not shown in the drawings, the user device according to the present invention may further be provided with an application chip set, a camera image processor (CIS), a mobile DRAM, and the like. Self-explanatory to those who have learned.

The storage device according to the present invention may be mounted using various types of packages. For example, flash memory and / or controllers may include Package on Package (PoP), Ball grid arrays (BGAs), Chip scale packages (CSPs), Plastic Leaded Chip Carrier (PLCC), Plastic Dual In-Line Package (PDIP), Die in Waffle Pack, Die in Wafer Form, Chip On Board (COB), Ceramic Dual In-Line Package (CERDIP), Plastic Metric Quad Flat Pack (MQFP), Thin Quad Flatpack (TQFP), Small Outline (SOIC), Shrink Small Outline Package (SSOP), Thin Small Outline (TSOP), Thin Quad Flatpack (TQFP), System In Package (SIP), Multi Chip Package (MCP), Wafer-level Fabricated Package (WFP), Wafer-Level Processed Stack Package It may be implemented using packages such as (WSP).

Meanwhile, in the detailed description of the present invention, specific embodiments have been described, but various modifications are possible without departing from the scope of the present invention. For example, the scope of the present invention is not limited to the flash memory device. The present invention can be applied to any storage device in which address translation by the translation layer is used. Therefore, the scope of the present invention should not be limited to the above-described embodiments, but should be defined by the equivalents of the claims of the present invention as well as the following claims.

1 is a block diagram illustrating an example of a user device performing an invalidation operation using a logging scheme.

FIG. 2 is a block diagram schematically illustrating the SSD controller shown in FIG. 1.

3 is a block diagram illustrating another embodiment of the SSD controller shown in FIG. 1.

4 is a flowchart illustrating an operation of a user device according to an exemplary embodiment of the present invention.

5 is a flowchart illustrating an operation of a user device according to another exemplary embodiment.

6 is a block diagram illustrating an embodiment of a user device for performing an invalidation operation using file system information.

FIG. 7 is a conceptual diagram illustrating a merge method of the user device illustrated in FIG. 6.

FIG. 8 is a flowchart illustrating an invalidation operation of the user device illustrated in FIG. 6.

9 is a block diagram illustrating an example of a user device for invalidating data in a buffer memory.

10 and 11 exemplarily illustrate mapping tables of the controller 3260 illustrated in FIG. 9.

FIG. 12 is a flowchart for describing an operation of invalidating buffer memory data of the user device illustrated in FIG. 9.

13 to 15 are diagrams for describing an invalidation operation of the user device illustrated in FIG. 9.

16 is a block diagram illustrating another embodiment of a user device for invalidating data in a buffer memory.

17 is a block diagram illustrating an example of a user device performing an invalidation operation using a file delete command.

18 is a diagram showing a data structure in accordance with the present invention that associates memory allocation units of a storage device with an indicator indicating whether memory allocation units contain valid or invalid data.

19 to 23 are flowcharts for describing operations of the user device illustrated in FIG. 17. .

24 is a block diagram illustrating a storage device that performs self invalidation operation without host intervention.

25 shows an example of logically partitioning a storage area of a nonvolatile storage medium.

FIG. 26 shows a well-known 512 byte example of the MBR shown in FIG. 25.

FIG. 27 shows an example of the layout of a single 16 byte partition record of the MBR shown in FIG. 26.

28 is a table showing partition types and corresponding ID values.

29 to 34 illustrate a method of invalidating an erased data area of a solid state drive (SSD) according to an exemplary embodiment of the present invention.

35 is a block diagram of a user device according to an embodiment of the present invention.

36 is a block diagram showing a software structure of a user device performing a recovery operation.

37 is a block diagram illustrating a hardware structure of a user device performing a recovery operation.

38 is a flowchart illustrating a data deletion operation.

39 is a conceptual view illustrating a method of deleting meta data when data is deleted. 39 (a) shows before the metadata is deleted, and FIG. 39 (b) shows after the metadata is deleted.

40 is a block diagram illustrating a method of invalidating mapping data corresponding to data to be deleted during a data deletion operation. 40 (a) shows a mapping table before data deletion, and FIG. 40 (b) shows a mapping table after data deletion.

41 is a conceptual diagram for explaining a merge operation. FIG. 41A shows the mapping table before merge and FIG. 41B shows the mapping table after merge.

42 is a conceptual diagram for explaining a method of operating an invalid delay queue.

FIG. 43 is a flowchart illustrating a data recovery method using the invalid delay queue of FIG. 42.

44 is a conceptual diagram illustrating a method of operating a merge / erase prevention queue.

45 is a flowchart illustrating a data recovery method using the merge / erase prevention queue of FIG. 44.

46 is a conceptual diagram for explaining an operation method when an invalid delay queue and a merge / erase prevention queue are used at the same time. 46 (a) shows a case where only an invalid delay queue is used, and FIG. 46 (b) shows a case where an invalid delay queue and a merge / erase prevention queue are used together.

FIG. 47 is a flowchart illustrating a data recovery method using both an invalid delay queue and a merge / erase prevention queue of FIG. 46.

48 is a block diagram schematically illustrating a user device including a semiconductor memory device according to an embodiment of the present invention.

Claims (22)

A storage medium for storing data; And A controller for controlling the storage medium, And the controller selectively performs an invalidation operation or a logging operation according to whether the size of data to be deleted exceeds a reference size. The method of claim 1, The reference size may be variable through a firmware update. The method of claim 1, And the controller performs an invalidation operation without delay when the data to be deleted requires security. The method of claim 1, The controller is configured to perform an invalidation operation using file system information. The method of claim 4, wherein The storage medium includes an area for storing the file system information. The method of claim 1, A buffer memory for temporarily storing data to be stored in the storage medium, And the controller invalidates data stored in the buffer memory in response to an invalidation command. The method of claim 6, The controller includes a mapping table for invalidating data stored in the buffer memory. The method of claim 7, wherein And the controller updates the mapping table in response to the invalidation command. The method of claim 1, And when the write operation is requested, the controller determines whether the write operation is related to metadata, and updates mapping information of user data according to the determination result. The method of claim 9, And when it is determined that the write operation is related to partition metadata, the controller updates the mapping information so that user data corresponding to the partition metadata is invalidated. The method of claim 1, And the controller invalidates invalid data stored in the storage medium in response to an invalidation command. The method of claim 1, And the controller delays an erase operation on the data to be deleted. 13. The method of claim 12, The controller includes an invalid delay queue for delaying an invalidation operation on data to be deleted. The method of claim 13, And data to be deleted are sequentially registered in the invalid delay queue, and are sequentially invalidated according to a first-in first-out method. A host providing an invalidation command; And Including a storage device for performing an invalidation operation according to the invalidation command, The storage device, A storage medium for storing data; And And a controller which controls the storage medium and selectively performs an invalidation operation or a logging operation according to whether the size of data to be deleted exceeds a reference size. The method of claim 15, And the controller performs an invalidation operation without delay when data to be deleted requires security. The method of claim 15, The storage device further includes a buffer memory for temporarily storing data to be stored in the storage medium, And the controller invalidates data stored in the buffer memory in response to the invalidation command. The method of claim 17, And the controller includes a mapping table for invalidating data stored in the buffer memory. The method of claim 18, The controller is configured to update the mapping table in response to the invalidation command. The method of claim 15, And the controller invalidates invalid data stored in the storage medium in response to an invalidation command. The method of claim 15, The controller further comprises an invalid delay queue for delaying an invalidation operation on data to be deleted. The method of claim 21, And data to be deleted are sequentially registered in the invalid delay queue, and are invalidated sequentially according to a first-in first-out method.

Images