KR101541532B1 - Method and apparatus for data consistency control of storage class memory - Google Patents

Abstract

The present invention relates to a computing device including a storage class memory. According to an embodiment of the present invention, the device comprises a storage class memory storing a program which operates a file system; and a processor for operating the program stored in the storage class memory, wherein the processor records a log history of data stored in a log area of the storage class memory when an update request for updating the data stored in the storage class memory is created by execution of the program, and updates the data according to the update request in the storage class memory based on a log type.

Description

[0001] METHOD AND APPARATUS FOR DATA CONSISTENCY CONTROL OF STORAGE CLASS MEMORY [0002]

The present invention relates to an apparatus and method for maintaining data consistency of a storage class memory.

In a computing system, the memory hierarchy is a memory partition that depends on the speed at which the CPU can access. The hierarchical structure of the memory has a faster access speed as it goes to a higher level, but has a relatively small memory capacity because the price per unit of capacity is high. On the other hand, the access speed is slower toward the lower level, and the price per capacity unit is relatively low, so that the memory capacity is relatively large.

Secondary storage belonging to a lower level in the hierarchical structure of memory has a slow access speed of the CPU, but is inexpensive and has a larger storage capacity than other memories. In addition, since the auxiliary storage device is non-volatile, data can be semi-permanently stored. The main storage, which is higher than the auxiliary storage, has a faster access speed than the auxiliary storage but is more expensive than the auxiliary storage, has less storage capacity, and is volatile. Therefore, the auxiliary storage device is used as a storage device for assisting the main storage device.

At this time, the auxiliary storage device is collectively referred to as a storage device which keeps stored information even when power is not supplied. For example, the auxiliary storage device may be a compact flash (CF) card, a secure digital (SD) card, a memory stick, a solid-state drive (SSD) Magnetic disk storage devices such as NAND flash memory, hard disk drive (HDD), and the like, and optical disk drives such as CD-ROM, DVD-ROM, etc. have.

Also, the main storage device is collectively referred to as a memory that requires electricity to maintain stored information. For example, the main memory may be a fast page mode DRAM (FPM), an extended data out (DRAM) DRAM, a synchronous DRAM (SDRAM), a double data rate SDRAM (DDR SDRAM), a double data rate 2 SDRAM (DDR2 SDRAM) , Dynamic random access memory (DRAM) such as double data rate 3 (SDRAM), static random access memory (SRAM) such as bipolar SRAM and CMOS SRAM, and the like.

The auxiliary storage device is cheaper than the main storage device at a price per capacity unit, and is characterized by nonvolatility, so that data can be semi-permanently recorded. However, the auxiliary memory has a drawback that it is slower to read and write data than the upper memory of the memory hierarchy. Therefore, the auxiliary storage device having such disadvantages may cause degradation of performance of the entire computing system. To overcome this drawback, the operating system's file system uses a portion of the main memory as a page cache. However, the page cache manages data in both main memory and auxiliary memory, which can cause data consistency maintenance problems.

Recently, a storage class memory has been newly developed to solve the disadvantages of the main storage device and the auxiliary storage device. The storage class memory is a nonvolatile memory that simultaneously provides attributes of the nonvolatile memory and attributes of the volatile memory. Storage class memory is capable of high-speed byte-by-byte random access and has a fast access speed. For example, the storage class memory may include phase-change RAM (PCM), ferroelectric RAM (FeRAM), magnetic RAM (MRAM), resistive RAM (RRAM), STT-MRAM, and the like.

Storage class memory, which provides both nonvolatile and volatile attributes, and supports high-speed byte-by-byte random access, can be used as an alternative to existing auxiliary storage or main storage. Also, it can be used as a part of existing auxiliary memory for storing operating system or frequently used files, and can be utilized as part of main memory for programs that are continuously used even when power is cut off by using nonvolatile property. The storage class memory can be utilized as a kind of write buffer cache space between the main memory and the auxiliary memory.

However, there are some considerations for using storage class memory as auxiliary storage. As mentioned above, the file system for existing auxiliary storage supports block-level access and solves the performance degradation of the whole computing system due to the slow access speed by using a part of main memory as the page cache. However, storage class memory is accessed byte by byte, unlike conventional auxiliary memory units that access block by block. Therefore, when using an existing file system in storage class memory, a different method of managing the storage is required. And because the storage class memory is as fast as the main memory, using the page cache to reduce the speed difference between the storage and the main memory may cause the data write performance to be reduced by managing the data redundantly for no reason.

In addition, since the storage class memory supports byte-by-byte access, it needs to be synchronized with the cache of the CPU in order to maintain cache coherency because it shares the memory bus with the main memory. Computing systems that use storage class memory use the "cache flush" command for this purpose. The "cache flush" command is a command to update cache data in memory when it is loaded into the cache and inconsistency exists between data in use and data in memory. However, in the process of using "cache flush", log data is generated, which increases the amount of writing and caching of log data. Therefore, there is a need for a file system for considering the characteristics of such a storage class memory and for maintaining data consistency.

The existing invention for maintaining the consistency of the auxiliary storage device in the file system is as follows.

Korean Patent Registration No. 0501414 entitled "Logging and Recovery Method and Apparatus for Recovery of Metadata of File System") uses a log of already stored metadata to maintain consistency of metadata in the file system Thereby performing log recording and recovery for recovery of metadata.

Korean Patent Publication No. 2012-0029239 entitled "Data Recording System and Method Using PRAM") relates to a data recording system and a method thereof using a phase-change RAM (PRAM). The present invention discloses a method of maintaining system consistency by supporting block atomicity in a system using PRAM, which is a nonvolatile memory supporting block writing, as a data storage device.

SUMMARY OF THE INVENTION It is an object of the present invention to provide an apparatus and method for maintaining data consistency in a storage class memory in order to reduce the size of log data and minimize the cost of consistency. have.

It should be understood, however, that the technical scope of the present invention is not limited to the above-described technical problems, and other technical problems may exist.

According to an aspect of the present invention, a computing device including a storage class memory according to an embodiment of the present invention includes a storage class memory in which a program for operating a file system is stored, Processor. At this time, when an update request for updating the data stored in the storage class memory occurs, the processor records a log of data stored in the log area included in the storage class memory according to the execution of the program, And updates the data to be updated according to the update request in the class memory.

According to another aspect of the present invention, there is provided a method for maintaining data consistency in a file system based on a storage class memory, comprising: generating an update request for updating data stored in a storage class memory; Recording a log of data stored in a log area included in the storage class memory; And updating the data to be updated according to the update request in the storage class memory based on the type of the log.

According to any one of the above-mentioned problems, the present invention can provide a data consistency maintaining apparatus and method for supporting a storage class memory having both advantages of a volatile memory and a nonvolatile memory, The ripple effects can be expected.

The apparatus and method for maintaining data consistency of the storage class memory according to the present invention can reduce unnecessary backup cost compared with the existing file system and reduce the amount of data of a file to be recorded in the storage class memory, It can provide faster system performance than existing file systems.

1 is an exemplary diagram for explaining log data of a file system using an existing WAL.
2 is a schematic diagram illustrating a computing device including a storage class memory according to an embodiment of the present invention.
FIG. 3 is an exemplary diagram illustrating a process of recording a log and a process of updating data according to an embodiment of the present invention. Referring to FIG.
4 is a block diagram schematically illustrating a storage class memory including a log area, a first data block, and a second data block according to an embodiment of the present invention.
5 is an exemplary diagram illustrating a data recovery process according to an embodiment of the present invention.
6 is a flowchart illustrating a method of maintaining data consistency in a file system based on a storage class memory according to an embodiment of the present invention.
FIG. 7 is a flowchart illustrating a method of recording logs in a log area in a method of maintaining data consistency in a file system according to an exemplary embodiment of the present invention. Referring to FIG.
8 is a flowchart illustrating a data updating step in a data consistency maintenance method of a file system according to an embodiment of the present invention.
FIG. 9 is a flowchart illustrating a method of recovering data using a log in a method of maintaining data consistency in a file system according to an exemplary embodiment of the present invention. Referring to FIG.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, which will be readily apparent to those skilled in the art. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly explain the present invention in the drawings, parts not related to the description are omitted.

Throughout the specification, when a part is referred to as being "connected" to another part, it includes not only "directly connected" but also "electrically connected" with another part in between . Also, when a part is referred to as "including " an element, it does not exclude other elements unless specifically stated otherwise.

In the following description with reference to the drawings, the same reference numerals denote the same reference numerals as in the drawings, and the reference numerals are merely described for convenience of description. Therefore, the concept, feature, function, or effect of each constitution is not limited by the corresponding drawing number.

1 is an exemplary diagram for explaining log data of a file system using existing write-ahead logging (WAL).

Referring to FIG. 1, a data block in a file system is a space for storing an actual file. In this case, an inode is an abbreviation of an index node, and stores metadata about a file or a directory. In addition, the inode may include a direct block that stores pointers to data blocks. However, in practice, a direct block can not store all the pointers to data blocks contained in a storage device, so an indirect block, which is a dynamic block, is used when more data blocks are needed. An indirect block stores a pointer to a block that stores a pointer to a data block. The indirect block includes a single indirect block, a double indirect block, and a triple indirect block.

The log area shown at the bottom of FIG. 1 is an area for recording logs in the file system of the storage device. File systems that use existing WALs log to the log area to store new data in the storage device or to update pre-stored data. At this time, the information to be recorded may include all data newly stored or updated in the inode and data block of the storage device. Therefore, as shown in FIG. 1, in the log area, 8 bytes of data stored in the inode and 3 kilobytes, 4 kilobytes, and 1.5 kilobytes of data stored in the three data blocks are recorded. That is, a total of 8,712 bytes of the log is recorded in the log area.

The following is a description of a computing device including a storage class memory in accordance with one embodiment of the present invention with reference to FIG. 2 is a schematic diagram illustrating a computing device including a storage class memory according to an embodiment of the present invention.

The computing device 10 including the storage class memory 300 according to an embodiment of the present invention includes a storage class memory 300 in which a program for operating a file system is stored and a storage class memory 300 in which a program stored in the storage class memory 300 is operated Processor < / RTI >

Meanwhile, the computing device 10 according to an exemplary embodiment of the present invention may include a general computer such as a server, a workstation, a desktop computer and a notebook computer, a smartphone, Tablet PCs, and smart devices such as smart TVs.

The storage class memory 300 also includes a memory that provides non-volatile attributes such as phase change RAM (PCM), ferroelectric RAM (FeRAM), magnetic RAM (MRAM), resistive RAM (RRAM) .

The computing device 10 may further include a memory 200 and a storage device 400 according to an embodiment of the present invention.

At this time, the memory is collectively referred to as a volatile memory. For example, the memory 200 may be a memory such as a fast page mode DRAM (FPM), an extended data out (DRAM), an synchronous DRAM, a double data rate (SDRAM) ), Dynamic random access memory (DRAM) such as double data rate 3 (SDRAM), static random access memory (SRAM) such as bipolar SRAM and CMOS SRAM, and the like.

In addition, the storage device 400 collectively refers to a non-volatile storage device that keeps stored information even when power is not supplied. For example, the storage device 400 may be a compact flash (CF) card, a secure digital (SD) card, a memory stick, a solid-state drive (SSD) A magnetic computer storage device such as a NAND flash memory, a hard disk drive (HDD) and the like, an optical disc drive such as a CD-ROM, a DVD-ROM, etc., can do.

The storage class memory 300 can be utilized as an expansion of the storage device 400 or a write buffer cache space of the memory 200 and the storage device 400 using a fast access speed in the computing system 10. In addition, the storage class memory 300 has non-volatile attributes and volatile attributes, so that it can be used in place of the memory 200 or the storage device 400 in the computing system 10. Thus, in another embodiment of the present invention, the storage class memory 300 may be utilized as the memory 200 of the computer device 10. [ Also, in another embodiment of the present invention, the storage class memory 300 may be utilized as the storage device 400 of the computing device 10.

In addition, the storage class memory 300 may include a log area for recording logs and a data area for storing data. At this time, the data area may include a file or a data block storing data.

The program stored in the storage class memory 300 according to an exemplary embodiment of the present invention may operate as an operating system for operating a file system to which an embodiment of the present invention is applied or a file system to which an embodiment of the present invention is applied And can be configured in the form of an application loaded.

The file system refers to a system for storing and managing files and data in a storage device 400 and a database connected to the computing device 10. For example, the file system may include a file allocation table 32 (FAT32), a new technology file system (NTFS), an extended file system 3 (EXT3), and an extended file system 4 (EXT4).

The processor 100 according to an embodiment of the present invention executes a program in response to a request to update the changed data in the storage class memory 300, Record the log. And performs data update on the storage class memory 300 based on the type of log.

At this time, the update request for the changed data in the storage class memory 300 occurs when the data stored in the cache memory or the memory 200 included in the processor 100 or the storage class memory 300 is inconsistent. That is, the update request may be used to maintain data consistency by reflecting the changed data in the storage class memory 300. [ In response to such an update request, the processor 100 may record the log in the log area included in the storage class memory 300. [

Next, a process of writing logs and updating data in the storage class memory 300 according to an embodiment of the present invention will be described with reference to FIGS. 3 and 4. FIG. FIG. 3 is a diagram for explaining a log recording process and a data update process according to an embodiment of the present invention. FIG. 4 is a diagram illustrating a log area 310, And a storage class memory 300 including a block 320 and a second data block 330. As shown in FIG.

The processor 100 according to an exemplary embodiment of the present invention may increase the number of times the cache flush command is executed as well as the write amount when the number of logs recorded in the log area 310 included in the storage class memory 300 increases. Therefore, the processor 100 needs to minimize the log recorded in the log area 310.

To this end, an embodiment of the present invention may set the log type differently according to the ratio of data updated in the data block of the storage class memory 300. A different log may be recorded in the log area 310 according to the type of the log.

The processor 100 may calculate the data rate to be updated in the data block of the storage class memory 300 in order to record the log in the log area 310. [ The processor 100 may then determine the type of log according to the updated data rate. At this time, the type of the log may include a data log and a pointer log.

When the data rate updated in the storage class memory 300 is less than the preset ratio, the amount of data recorded in the log area 310 may be relatively small even if the data to be updated is stored in the log area 310 as it is. Therefore, when the ratio is less than the preset ratio, the processor 100 can record the data log in the log area 310. [ At this time, the data log may include data stored in the storage class memory 300 to be updated.

On the contrary, when the updated data rate is equal to or greater than a predetermined rate, the amount of data recorded in the log area 310 may be relatively large if the data to be updated is stored in the log area 310 as it is. Therefore, when the predetermined ratio is equal to or higher than the preset ratio, the processor 100 can record the pointer log in the log area 310. [ At this time, the pointer log is a log including pointers to data blocks of the storage class memory 300 in which data to be updated is stored.

As described above, according to an embodiment of the present invention, log data can be minimized by setting different log types according to the changed data rate.

As shown in FIG. 3, when there are a plurality of data blocks subject to the update request, the data ratios updated in the respective data blocks are calculated, and pointer logs and data logs corresponding to the respective data blocks are stored in the log areas 310 ). ≪ / RTI > In one embodiment of the present invention, the size of the log data can be minimized as shown in the lower part of FIG. 3 by recording the log. At this time, the size of the log data to be stored is 1,048 bytes, which occupies 1/8 less capacity than the existing WAL method in FIG.

The predetermined ratio for distinguishing the type of log may vary in the computing system 10 depending on the usage environment. For example, in one embodiment of the present invention, the predetermined ratio can be set to 50%.

Referring to FIG. 4, for example, an update request for data stored in the first data block 320 may occur in the storage class memory 300 including the first data block 320 and the second data block. The processor 100 may calculate the data rate to be changed in the first data block 320 in response to the update request. If the changed data rate is less than 50%, which is a predetermined ratio, the processor 100 writes a data log including data to be updated in the log area 310, among the data stored in the first data block 320 . If the changed data rate is 50% or more, the processor 100 may write a pointer log including the pointer of the first data block 320 in the log area 310. [

A state in which all logs for each data block to be changed are recorded in the log area 310 of the storage class memory 300 is called a commit state. Once committed, the processor 100 may update the data in the storage class memory 300 based on the type of log.

After the commit state is established, the processor 100 may first determine the type of log recorded in the log area according to the log recording method described above in order to update the data. At this time, when the type of log associated with the data block to be updated is the data log-in, the processor 100 can update the data directly to the data block to be updated. Alternatively, if the type of log is a pointer login, the processor 100 may update the data in a data block different from the data block to be updated.

4, if the type of log associated with the first data block 320 stored in the storage class memory 300 is a data logon, the processor 100 updates the data in the first data block 320 can do. Alternatively, if the type of log associated with the first data block 320 is a pointer login, the processor 100 may update the data in a second data block 330 that is different from the first data block 320.

The processor 100 updates the storage class memory 300 with data to be updated according to the update request, and when the data update is successfully completed, the processor 100 deletes the log using the check-out process, Can be completed.

Next, a recovery process of the storage class memory 300 according to an embodiment of the present invention will be described with reference to FIGS. 4 and 5. FIG. 5 is an exemplary diagram illustrating a data recovery process according to an embodiment of the present invention.

The processor 100 according to an embodiment of the present invention can recover data using a log recorded in the log area 310 when an error occurs while updating data in the storage class memory 300. [

Here, an error means that a system failure occurs while updating the data in the computing device 10. The system failure may occur due to a power failure of the computing device 10 or a system error, but is not limited thereto. When the system failure occurs, the computing device 10 can not update the data in the storage class memory 300 any more.

When the system fails and the data can not be updated, the processor 100 according to an embodiment of the present invention can perform data recovery using the log recorded in the log area 310. [ To do this, the processor 100 may first determine the type of log associated with the data block to be recovered. At this time, if the log type is a data log-in, the processor 100 can recover the data of the data block using the data log. On the other hand, when the log type is a pointer log, the processor 100 can recover data using pointer information of a data block included in the pointer log. After the processor 100 performs the recovery using the pointer log, the processor 100 may delete the data block storing the updated data.

That is, when a system failure occurs while writing updated data as shown in FIG. 5, an embodiment of the present invention rolls back the data before the change using the log recorded in the log area 310 in advance, Consistency can be ensured.

As described above, according to an embodiment of the present invention, a recovery method may be different depending on the type of log associated with a data block to be restored. If the type of the log is a data log, the processor 100 may perform recovery using data recorded in the log area 310 as shown in FIG. 5A. If the type of the log is a pointer, the processor 100 uses the pointer information recorded in the log area 310 as shown in FIG. 5B to store the data block in which the current data is stored in the data block in which the previous data was stored To recover the data. At this time, after recovering the data using the pointer information, the data block storing the updated data can be deleted and the recovery can be completed.

For example, referring to FIG. 4, if a system failure occurs during data update according to an embodiment of the present invention, the processor 100 may determine the type of log recorded in the log area 310 have. If a data log of the first data block is recorded in the log area 310, the processor 100 may recover the data of the first data block using the data log. If a pointer log for storing pointer information of the first data block is recorded in the log area 310, the processor 100 uses the pointer information of the first data block to convert the second data block, Data blocks can be replaced. After performing the recovery using the pointer log, the data can be deleted in the second data block newly allocated when the data is stored.

Next, a method of maintaining data consistency in a storage system based on the storage class memory 300 according to an embodiment of the present invention will be described with reference to FIGS. 6 to 9. FIG.

6 is a flowchart illustrating a method of maintaining data consistency in a file system based on the storage class memory 300 according to an embodiment of the present invention.

The method of maintaining data consistency in a file system based on the storage class memory 300 according to an embodiment of the present invention is a method of maintaining data consistency in a storage class memory 300 when a update request for updating data stored in the storage class memory 300 is generated A log of data stored in the log area 310 included in the storage 300 is updated in step S510 and updated in the storage class memory 300 based on the type of log with data to be updated according to the update request in step S520. .

At this time, the storage class memory 300 is a memory having both a nonvolatile property and a volatile property, such as a phase change RAM (PCM), a ferroelectric RAM (FeRAM), a magnetic RAM (MRAM), a resistive RAM (RRAM) . ≪ / RTI > In addition, the storage class memory 300 may include a log area 310 for recording logs and a data area for storing data. At this time, the data area may include one or more data blocks 320 and 330 for storing files or data.

The file system may refer to a system for storing and managing files and data of the storage class memory 300. For example, the file system may include a file allocation table 32 (FAT32), a new technology file system (NTFS), an extended file system 3 (EXT3), and an extended file system 4 (EXT4).

In an embodiment of the present invention, the update request may be issued to the storage class memory 300 when the data stored in the cache memory or the memory 200 included in the processor 100 and the data stored in the storage class memory 300 are inconsistent. And can be used to maintain data consistency. According to an update request, an embodiment of the present invention may record a log in the log area 310 included in the storage class memory 300. [

At this time, the type of log stored in the log area 310 may include a data log and a pointer log. The data log may be used when the amount of data recorded in the log area 310 is relatively small and may include data stored in the storage class memory 300 to be updated. The pointer log may be used when a relatively large amount of data is recorded in the log area 310 and may include a pointer to a data block in which data to be updated is stored.

7 is a flowchart illustrating a process of recording a log in the log area 310 in the method of maintaining data consistency in a file system according to an embodiment of the present invention.

First, the file system may calculate the data rate to be changed in each data block of the storage class memory 300 to record the log of the data stored in the log area 310 of the storage class memory 300 (S510) ). The file system can determine whether the changed data rate of the data block is equal to or greater than a preset ratio in order to record the log in the log area 310 using the thus calculated ratio (S510). At this time, if the changed data rate is equal to or greater than a preset ratio, a pointer log including a pointer of a data block storing updated data in the log area 310 may be recorded (S514). Alternatively, if the changed data rate is less than the preset ratio, the data log including the updated data may be recorded in the log area 310 (S513).

The predetermined rate for classifying the log may vary depending on the environment in which the computing system 10 including the storage class memory 300 is executed. For example, one embodiment of the present invention can set 50% at a predetermined rate. At this time, if the changed data rate is 50% or more, the pointer log is recorded (S514). If the changed data rate is less than 50%, the data log may be recorded (S513).

8 is a flowchart illustrating a data updating step in a data consistency maintenance method of a file system according to an embodiment of the present invention.

In order to update the data (S520), the file system may first determine the type of log of the data block that needs to update the data (S521, S522). If the type of the log is the data log, the data can be updated in the data block in which the data is stored (S523). On the other hand, if the log type is a pointer, the data can be stored in a data block different from the data block in which the data to be updated is stored (S524).

After the data update (S520) is successful, the log recorded in the log area 310 may be deleted using the checkout process to complete the data update (S525, S527).

If an error occurs while the file system updates the data in the storage class memory (S520), the file system can recover the data using the log (S526). At this time, the error may include all kinds of system failures such as a power failure of the computing device and a system failure. In an embodiment of the present invention, when a system failure occurs, the log data recorded in the log area 310 may be used to roll back the data before the change to ensure the consistency of the data.

FIG. 9 is a flowchart illustrating a method of recovering data using a log in a method of maintaining data consistency in a file system according to an exemplary embodiment of the present invention. Referring to FIG.

In order to restore using the log in the storage class memory 300 (S540), the file system may determine the type of log of the data block from which data is to be recovered (S541). At this time, if the log type of the data block is a data log, the data of the data block can be recovered using the information stored in the data log (S542). Alternatively, if the log type of the data block is a pointer, the data of the data block may be recovered using the pointer information included in the pointer log (S543), and the newly allocated data block may be deleted to store the data S545).

One embodiment of the present invention may also be embodied in the form of a recording medium including instructions executable by a computer, such as program modules, being executed by a computer. Computer readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media. In addition, the computer-readable medium may include both computer storage media and communication media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Communication media typically includes any information delivery media, including computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave, or other transport mechanism.

While the methods and systems of the present invention have been described in connection with specific embodiments, some or all of those elements or operations may be implemented using a computer system having a general purpose hardware architecture.

It will be understood by those skilled in the art that the foregoing description of the present invention is for illustrative purposes only and that those of ordinary skill in the art can readily understand that various changes and modifications may be made without departing from the spirit or essential characteristics of the present invention. will be. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive. For example, each component described as a single entity may be distributed and implemented, and components described as being distributed may also be implemented in a combined form.

The scope of the present invention is defined by the appended claims rather than the detailed description and all changes or modifications derived from the meaning and scope of the claims and their equivalents are to be construed as being included within the scope of the present invention do.

10: A computing device comprising storage class memory
100: Processor
200: memory
300: Storage class memory
400: storage device

Claims (12)

A computing device comprising a storage class memory,
The storage class memory in which the programs that run the file system are stored, and
And a processor for operating a program stored in the storage class memory,
Wherein the processor determines the type of the log based on the data requested to be updated when an update request for updating the data stored in the storage class memory occurs according to the execution of the program, Generates a log for the data,
Updating the data requested to be updated in the storage class memory based on the type of the log recorded after recording the generated log in a log area of the storage class memory,
After completion of the update of the data, deleting the log using the check-out processor to complete the data update,
Wherein the type of the log includes a data log for recording the data and a point log for recording points of the data. The method according to claim 1,
Wherein the processor is further configured to determine, based on a ratio of data updated in a data block of the storage class memory,
Generating the data log including the stored data in the log area if the ratio of the updated data is less than a predetermined ratio,
And generates the pointer log including a pointer to a data block in which the stored data is stored in the log area if the ratio of the updated data is equal to or greater than a predetermined ratio. The method according to claim 1,
The processor determines a type of the log to update the data stored in the storage class memory based on the type of the log,
Updating the data in the data block of the storage class memory if the type of the log is a data log,
And updates the data in a data block different from the data block if the type of log is a pointer login. The method according to claim 1,
Wherein the processor recovers the data using the log when an error occurs in updating the data in the storage class memory. 5. The method of claim 4,
Wherein the processor determines the type of the log to recover the data using the log, and if the type of the log is a data log, restores the data of the data block using the data log,
Wherein when the type of the log is a pointer, the data of the data block is recovered by using pointer information of the data block included in the pointer log, and the data block stored with data after the update is deleted, . delete A method for maintaining data consistency in a storage class memory based file system,
Generating an update request to update data stored in the storage class memory;
Determining a type of the log based on the data for which the update is requested;
Generating a log for the data according to the determined type of log;
Recording the generated log in a log area included in the storage class memory;
Updating data in the storage class memory based on the type of the logged log, the data being updated in accordance with the update request; And
And deleting the log using the check-out processor after completing the data update to complete the data update,
Wherein the type of the log includes a data log for recording the data and a point log for recording points of the data. 8. The method of claim 7,
Wherein the generating the log comprises:
And calculating a changed data rate of the data block in which the data is stored in the storage class memory,
Generating the data log including the updated data if the ratio of the updated data is less than a predetermined ratio,
Wherein the pointer log includes a pointer of a data block in which the updated data is stored if the ratio of the updated data is equal to or greater than a predetermined ratio. 8. The method of claim 7,
The step of updating the data comprises:
Determining a type of log of the data block in which the data is stored in the storage class memory;
Updating data in the data block in which the data is stored if the type of the log is a data log-in; And
Storing data in a data block different from the data block in which the data is stored when the type of the log is a pointer login. delete 8. The method of claim 7,
Generating an error in updating the data in the storage class memory; And
And recovering the erroneous data using the log. 12. The method of claim 11,
The step of recovering the error-
Determining a type of log of a data block corresponding to the data in which the error occurred;
Recovering data of the data block using information recorded in the data log when the type of log of the data block is a data log-in; And
If the type of log of the data block is a pointer login, restores the data of the data block using the pointer information of the data block included in the pointer log, and stores the newly allocated data block The method comprising the steps of:

Images