KR101969799B1 - Electronic device and controlling method thereof - Google Patents

Abstract

An electronic device and a control method are disclosed. The present disclosure includes a new memory including a main memory area and a file system area and a processor for performing an operation and performing checkpointing on a main memory area and a file system area according to predetermined conditions, The processor includes a checkpoint table for writing the changed page information. When the processor performs checkpointing on the main memory area, the processor initializes the checkpoint table and reads the attribute of the page in the main memory area as read-only (read-only).

Description

[0001] DESCRIPTION [0002] ELECTRONIC DEVICE AND CONTROL METHOD [0003]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electronic apparatus and a control method, and more particularly, to an electronic apparatus and a control method for managing data using check pointing.

Checkpointing is a commonly used mechanism for storing consistent state that can be returned by the system in the event of an unexpected error event. The checkpointing technique converts the state of the system to a file, which is stored on the storage device.

Referring to FIG. 1, a method of managing data using conventional checkpointing is illustrated. The checkpoint converts the system state to a file, and the converted file is stored in the storage device. If an error occurs during an operation, the electronic device loses time w because it must recover and restart the system at the most recent checkpointing. Also, it takes time r, which is the time required to restore the system at the checkpoint. Therefore, the conventional electronic device took time as long as w + r until recovery after occurrence of error. Generally, in a High Performance Computing (HPC) system, the continuous operation time (or checkpoint interval) is about 3 hours to 4 hours, and the check pointing time is about 30 minutes to 1 hour. Therefore, the conventional electronic device has a disadvantage in that it takes much time when an error occurs.

 In addition, conventional electronic devices require a complicated software management structure to ensure consistency and persistence of data because segment information, processor volatility information, and devices must all be stored synchronously.

SUMMARY OF THE INVENTION The present disclosure is directed to solve the above problems and an object of the present disclosure is to provide an electronic device and a control method that can provide efficient data consistency and persistence in order to secure reliability of system operation.

According to an embodiment of the present disclosure, there is provided a method for performing a new memory and operation including a main memory area and a file system area, and performing a new operation on the main memory area and the file system area Wherein the main memory area includes a checkpoint table for writing page information whose data has been changed, and when the processor performs checkpointing on a main memory area, And initializes a checkpoint table and sets an attribute of a page of the main memory area as read-only.

The processor may write the information of the first page to the checkpoint table when a data update of the first page is requested when the operation is performed.

The processor generates a second page in which the first page is copied, writes the information of the second page in the checkpoint table, changes the attribute to write, Can be updated.

The processor may also roll back the first page based on the information written to the checkpoint table if an error occurs when performing the operation or performing checkpointing on the main memory area .

Meanwhile, the file system area includes a buffer and a file storage unit, the file storage unit includes a previously stored data block, and the processor can store the updated data block in the buffer when the data block is updated .

The processor may write the updated data block to the file storage when performing checkpointing on the file system area.

In addition, the processor may perform checkpointing on the file system area after performing checkpointing on the main memory area.

The predetermined condition may be at least one of a predetermined period and a condition that the predetermined number of changes of the data is satisfied.

According to an embodiment of the present disclosure, there is provided a method for performing checkpointing, the method comprising: performing checkpointing and performing an operation, Check pointing is performed according to a predetermined condition for a main memory area of a new memory including a point table, check-pointing is performed on a file system area of the new memory, check-pointing is performed on the main memory area , And initializing the checkpoint table and setting an attribute of a page of the main memory area to be read-only.

And performing the operation may write the information of the first page to the checkpoint table when a data update of the first page is requested.

In addition, the step of performing the operation may include generating a second page in which the first page is copied, writing information of the second page in the checkpoint table, changing the attribute to write, One page of data can be updated.

In addition, the step of performing the operation may include, when an error occurs when performing the operation or performing checkpointing on the main memory area, performing a rollback of the first page based on the information written in the checkpoint table roll back.

The step of performing checkpointing may store the updated data block in the buffer when the data block previously stored in the file storage unit of the file system area including the buffer and the file storage unit is updated.

In addition, the checkpointing may write the updated data block to the file storage when performing checkpointing on the file system area.

The checking pointing may perform checkpointing on the file system area after performing checkpointing on the main memory area.

As described above, according to various embodiments of the present disclosure, electronic devices and control methods can increase the efficiency of data storage of data and ensure data consistency and persistence.

FIG. 1 is a diagram for explaining a process of performing conventional computation and checkpointing.
FIG. 2 is a diagram illustrating a process of performing computation and checkpointing according to an embodiment of the present disclosure.
3 is a block diagram of an electronic device according to one embodiment of the present disclosure;
4 is a detailed block diagram of a processor in accordance with one embodiment of the present disclosure;
5 is a diagram illustrating an architecture of an electronic device according to an embodiment of the present disclosure;
Figure 6 is a diagram illustrating a process by which an electronic device according to an embodiment of the present disclosure stores information about a process.
7 is a diagram illustrating a structure in which an electronic device according to an embodiment of the present disclosure manages information about a process.
8 is a diagram illustrating a process of storing data and metadata in an electronic device according to another embodiment of the present disclosure.
FIG. 9 is a diagram for explaining a process of performing computation and checkpointing according to an embodiment of the present disclosure.
10 is a flowchart of an electronic device control method according to an embodiment of the present disclosure.
11 is a flow diagram illustrating the operation of an electronic device that stores information regarding a particular process in accordance with an event in accordance with an embodiment of the present disclosure;
Figure 12 is a flow diagram illustrating the operation of an electronic device that stores specific data and metadata in accordance with an event in accordance with an embodiment of the present disclosure.

Various embodiments will now be described in detail with reference to the accompanying drawings. The embodiments described herein can be variously modified. Specific embodiments are described in the drawings and may be described in detail in the detailed description. It should be understood, however, that the specific embodiments disclosed in the accompanying drawings are intended only to facilitate understanding of various embodiments. Accordingly, it is to be understood that the technical idea is not limited by the specific embodiments disclosed in the accompanying drawings, but includes all equivalents or alternatives falling within the spirit and scope of the invention.

Terms including ordinals, such as first, second, etc., may be used to describe various elements, but such elements are not limited to the above terms. The above terms are used only for the purpose of distinguishing one component from another.

In this specification, the terms " comprises " or " having ", and the like, are intended to specify the presence of stated features, integers, steps, operations, elements, parts, or combinations thereof, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof. It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between.

In the meantime, " module " or " part " for components used in the present specification performs at least one function or operation. Also, " module " or " part " may perform functions or operations by hardware, software, or a combination of hardware and software. Also, a plurality of " modules " or a plurality of " parts ", other than a " module " or " part ", to be performed in a specific hardware or performed in at least one processor may be integrated into at least one module. The singular expressions include plural expressions unless the context clearly dictates otherwise.

In addition, in the description of the present invention, when it is judged that the detailed description of known functions or constructions related thereto may unnecessarily obscure the gist of the present invention, the detailed description thereof will be abbreviated or omitted.

FIG. 2 is a diagram illustrating a process of performing computation and checkpointing according to an embodiment of the present disclosure.

Referring to FIG. 2, an electronic device for performing an update in a file system area when a page fault occurs when performing computing is disclosed. The present disclosure uses new memory among non-volatile memories. New memory is a non-volatile memory that can be accessed on a byte-by-byte basis. For example, the new memory may include MRAM, FeRAM, RRAM, STT-MRAM, PRAM, 3D-XPoint, and the like. New memory can serve as conventional main memory and storage. Therefore, the electronic apparatus of the present disclosure has an advantage of not using the main memory and the storage separately. In addition, since the main memory and the storage are used as one new memory, there is an advantage in terms of the design space of the electronic device.

The new memory may include a main memory area and a file system area. The main memory area can perform the conventional main memory function used in the calculation process, and the file system area can perform the conventional storage function in which the final data is stored. When performing an operation, the electronic device can update only the data in the main memory area. Then, the electronic device can move the data of the main memory area to the file system area periodically, when a specific event occurs. For example, a specific event is when a page fault occurs. A page fault may mean that a page in the main memory area overflows or needs to be initialized according to certain conditions. The electronic device of the present disclosure moves the data in the main memory area to the file system area under certain conditions, so the checkpoint time can be reduced. In addition, since the main memory area and the file system area exist in one new memory, it may take a little time to move data from the main memory area to the file system area. In one embodiment, the continuous computing time (or checkpoint interval) of the present disclosure is about 10 seconds, and the page fault and checkpointing time for moving the data can be negligible when compared to the overall processing time of the electronic device It is time.

And, since the electronic device of this disclosure occasionally moves the data of the main memory area to the file system area, data for rolling back even if an error occurs is relatively small. Thus, the electronic device of the present disclosure can significantly reduce the rollback time compared to conventional electronic devices. In addition, the electronic device of the present disclosure can apply various schemes to reduce data movement time and ensure data persistency and consistency. A specific embodiment will be described later.

3 is a block diagram of an electronic device according to one embodiment of the present disclosure;

Referring to FIG. 3, the electronic device 100 includes a processor 110 and a memory 120. The memory 120 is a new memory including a main memory area and a file system area. As described above, the main memory area serves as a conventional main memory, and the file system area can perform a conventional storage function.

The main memory area of the memory 120 can update data that is changed during the operation. The main memory area may include a checkpoint table for writing page information whose data has been changed. The checkpoint table may include page information whose data has been changed or page information that has been copied. In one embodiment of the present disclosure, the storage unit of the memory 120 is described as a page, but the storage unit may be variously set as bytes, words, blocks, and the like.

And, the memory 120 includes a file system area. The file system area stores data of the main memory area according to a preset event or a preset period. The file system area includes a buffer and a file storage unit. The data is stored in a buffer while the electronic device 100 performs an operation. When the operation is terminated or an update signal is received, data in the buffer is written ). For example, when an operation is terminated or an update signal is received, it may be time to perform checkpointing on the file system area.

The processor 110 performs an operation and performs checkpointing on the main memory area and the file system area of the memory 120 according to predetermined conditions. When the processor 110 performs checkpointing on the main memory area, it can initialize the checkpoint table and set the property of the page in the main memory area to be read-only.

In addition, the processor 110 can organize data in the main memory area of the memory 120 and move the data to the file system area even during operation according to preset conditions. For example, the processor 110 may move data in the main memory area to the file system area when page faults occur, such as when the page is full of data, when the checkpoint table is full, and so on.

A detailed block diagram of the electronic device 100 is described below.

4 is a detailed block diagram of a processor in accordance with one embodiment of the present disclosure;

Referring to FIG. 4, the processor 110 may include a management unit 111, a control unit 112, and a restoration unit 113. In some cases, the processor 110 may include only some of the management unit 111, the control unit 112, or the restoration unit 113.

The management unit 111 may determine whether an event designated to perform checkpointing occurs. In one embodiment, the management unit 111 may determine, as an event, a context switch regarding the first process performed by the processor 110. [ In the following description, the context switching may indicate the operation of an electronic device that stores the state of a previously executed process to return to a new process state to switch the process performed by the processor. The first process may be an ongoing computation process.

In another embodiment, the management unit 111 may determine, as an event, a case in which the context switching of the first process performed by the processor 110 is performed a predetermined number of times.

The control unit 112 may store information on the first process at the time of the occurrence of the event in the new memory according to the determination result. More specifically, the control unit 112 may store the address relating to the first process as information for performing roll back. Further, the control unit 112 may copy the task structure related to the first process to the new memory using the fork () function call. The control unit 112 may update the page including the first data in the task structure in a CoW (Copy on Write) manner when a change of the first data with respect to the first process occurs.

In addition, the control unit 112 may store the latest address related to the first process as information for performing the rollback in the new memory, and delete the previously stored remaining address. In another embodiment, the control unit 112 may version the latest address for the first process. Accordingly, the control unit 112 can store a plurality of versions of a plurality of versions, each corresponding to a plurality of points in time, in the new memory.

In another embodiment, the control unit 112 may store the first data and the first metadata at the time when the event occurs in the file system.

The restoring unit 113 may delete the second data and the second metadata existing in the main memory area and roll back the first data and the first metadata when a system error related to the electronic device occurs. For example, the main memory area may be a process area. Further, the restoring unit 113 may restore the currently executing first process using the status information regarding the first process stored in the new memory.

On the other hand, the processor may include a manager for managing the memory. The processor may include a task manager for managing the main memory area and an IO manager for managing the file system area.

5 is a diagram illustrating an architecture of an electronic device according to an embodiment of the present disclosure;

5, the manager 10, the P-task 11, the P-io 12, the memory address space of the process 21, and the file of the process 22 Are shown. The manager 10 can manage the P-task 11 and the P-io 12 as a whole. The P-task 11 can manage processes such as calculation and check pointing of the main memory area as a task manager. The P-io 12 can manage processes such as calculation and check pointing of the file system area as an IO manager. The electronic device of the present disclosure can maintain data persistence by using a new memory including a main memory area and a file system area. Thus, in one embodiment, the task manager may be referred to as P (persistent) -task 11, and the IO manager may be referred to as P (persistent) -io.

The system state that must be stored in the memory for checkpointing may include information related to the workspace, i. E., The memory address space 21 of the process, the file 22 of the process. When storing the state of the memory address space and file space, both files can not be stored in any location, but they must be synchronized to the entire consistent state. The manager 10 synchronizes the data between the P-task 11 and the P-io 12 at each checkpointing period so that all data are consistent. For example, the memory address space 21 of the process may be a main memory area, and the file space may be a file system area.

Specifically, the P-task 11 (or the task manager) can maintain the consistency of the memory address space (or the main memory area) of the process. Updates to the memory address space occur only when the store header is executed. Therefore, the write operation differs from the file area (or the file system area) that the update operation calls. Thus, the P-task 11 makes a copy of the original data before making an update for every store instruction. In one embodiment, the memory can be managed on a page basis, so a copy is made on a page basis. However, no copy is generated from every store command. Unmodified raw data is data that will be rolled back due to transient errors.

The P-io 12 (or the IO manager) can maintain the consistency of the file system (or the file system area). The file system area may include a buffer and a file storage. That is, an update to a file occurs according to a write command for data or metadata. Like the P-task 11, the P-io 12 can alternately perform operations and checkpointing. A copy of the original data may be created before the update for all write commands in the operation step. However, P-io 12 may use a mechanism similar to managing the buffer cache in a traditional file system. When writing occurs, the modified block is stored in the buffer, and all reads and writes after this page can be processed in the buffer. However, the contents of the buffer are not removed and remain in the buffer until checkpointing is performed. The P-io 12 may flush all the pages of the buffer to the file storage unit at the checkpointing step.

The P-task 11 and the P-io 12 can independently check the work area (or the main memory area) of the process and the file of the process, respectively. The manager 10 synchronizes the P-task 11 and the P-io 12 to maintain data consistency. Then, the manager 10 performs a recovery process when an error occurs.

A method for the manager 10 to synchronize the P-task 11 and the P-io 12 will be described. The P-task 11 and the P-io 12 can alternately perform the operation and the checkpointing process. The operation of the main memory area and the operation of the file system area can be performed simultaneously. Checkpointing starts at a preset time. The checkpointing interval can be set by the user. The manager 10 stops the process associated with the application program. When a process associated with an application is stopped, the overall state of all processes can be synchronized. Next, the P-task checkpointing step is performed and then the P-io checkpointing step can be performed. The predefined checkpointing sequence allows the recovery process to return the application state in a consistent state when an error occurs.

Then, the manager 10 can perform the restoration process. When the error is restored, the electronics must return the application to a consistent state. The manager 10 can check the stored application program and perform the restoration process. If an error occurs in the computation step or the P-task checkpointing step, the application program can be restored to the previous consistent state. That is, the data can be rolled back. Alternatively, if an error occurs in the P-io checkpointing step, the data may be rolled forward since it has already passed the P-task checkpointing step. That is, in the electronic device of the present disclosure, since all the file data is in the buffer and writing data in the buffer to the file storage unit is idempotent, the consistency can be maintained by simply writing the buffer data in the file storage unit.

Figure 6 is a diagram illustrating a process by which an electronic device according to an embodiment of the present disclosure stores information about a process.

6 is an exemplary diagram illustrating a process by which an electronic device according to an embodiment stores information about a process. Referring to FIG. 6, an electronic device may be provided that generates and stores information P 1 and P 2 related to a process according to predetermined events 30, 40, and 50.

In this embodiment, the information about the process may indicate information at the time when a roll back is performed in preparation for a defect such as abnormal termination of the entire system. According to one embodiment, predetermined events 30, 40, 50 may be implemented with context switching.

As shown in FIG. 6, the context can be switched from the first process A1 to the second process A2 in accordance with the first context switch 30. Likewise, the context may be switched from the third process A3 back to the first process A1 according to the second context switch 30. However, the embodiments of the events 30, 40, and 50 shown in FIG. 6 are only exemplary descriptions for helping understanding of the present invention, and do not limit or limit the scope of other embodiments. For example, the point in time at which the context switching is performed a predetermined number of times may be used as an event. Specifically, the point in time when the command in which the context switching is executed is called n times can be used as an event.

The processor in the electronic device may newly perform the second process A2 in accordance with the first context switch 30. [ However, the electronic device of this embodiment can preliminarily store information P1 relating to the first process A1 for use as a rollback time in preparation for a defect such as a system error. More specifically, the electronic device can generate the information P1 about the first process A1 using the Fork () function. More specifically, the information P1 may indicate the address of the first process A1 at the time when the first context switch 30 occurs.

Further, the electronic device can generate the information P1 in the first process A1 in such a manner as to duplicate the task structure of the first process A1 that was being executed. Accordingly, the information P1 in the first process A1 and the first process A1 can be generated so as to share the same memory space.

There may be a case where various processes are executed by the electronic device and the first process A1 is executed again by the second context switching 40. [ The electronic device can independently update only the page information regarding the changed area from the information P1 relating to the first process A1.

Illustratively, if the first text 61 at the time the first context switch 30 was performed is stored as 2, then a change to the first text 61 occurs according to the second context switch 40 There may be cases. The electronic device can update the data 300 as the second text 62 in the Copy on Write manner for the first text 61. [ Thus, if a later third context switch 50 occurs, the electronic device determines that the information P2 about the first process A1 is the third text 63 (300) corresponding to the same value as the second text 62 . ≪ / RTI >

The electronic device according to the present embodiment can check-point the recent state of the process using the Fork () function and the Copy on Write method. Accordingly, the electronic device according to the present embodiment can be expected to have an effect of improving the durability by reducing the load of the entire system and using only light resources.

7 is a diagram illustrating a structure in which an electronic device according to an embodiment of the present disclosure manages information about a process.

Referring to FIG. 7, the new memory 120 included in the electronic device may include a run queue 121 and a persistence queue 122 for a running process. In one embodiment, run queue 121 and persistence queue 122 may be included in the main memory region of memory 120.

The electronic device may include a run queue 121 for a linked list in which processes running in the new memory 120 are stored. In this embodiment, the electronic device can manage the first process A1, the second process A2, and the third process A3 using the run queue 121. [

In addition, the electronic device may include a persistent queue 122 that stores information regarding processes that are not currently executing. Illustratively, in response to the first context switch 30, the information P1 about the first process A1 can be managed using the persistence turn queue 122. [ The information P1 relating to the first process A1 can indicate information on the state of the first process A1 at the time when the first context switch 30 has occurred. Illustratively, when a situation such as a system error occurs in the computing system, the point in time at which the first context switching 30 occurs can be used as a reference point for the process to be rolled back.

As an example, there may be a case where a system defect occurs while the first process A1 existing in the run queue 121 is being executed. In this case, the electronic device can forcibly terminate the first process A1 being executed. The electronic device may also copy the information P1 about the first process A1 previously stored in the persistent turn queue 122 to the run queue 121 and restore the first process A1 corresponding to the rollback time.

The electronic device according to the present embodiment can expect the effect of increasing the durability by storing the rollback information of the process while reducing the system load by utilizing the nonvolatile characteristic of NVRAMS which is widely used as a new memory today.

8 is a diagram illustrating a process of storing data and metadata in an electronic device according to another embodiment of the present disclosure.

Referring to FIG. 8, a structural diagram of a new memory 120 included in an electronic device according to the present embodiment is shown. The new memory 120 may include a main memory area 121 and a file system area 122.

The electronic device according to the present embodiment is configured to store the related data in a first meta existing in the main memory area 121 when a new data creation operation or a modification operation 531 is executed according to a running process, Data M1 can be updated. In addition, the electronic device can update the related data to the first data D1 existing in the main memory area 121 when the data writing operation 73a, 73b is executed according to the running process.

The electronic device can determine whether or not a predetermined event in which checkpointing is performed occurs. In one embodiment, the predefined event may represent a context change regarding the process performed by the electronic device. In another embodiment, the predefined event may indicate when the context switching of the process performed by the electronic device is performed a predetermined number of times.

The electronic device may copy 73a and 73b both the first data D1 and the first meta data M1 present in the main memory area 121 to the file system area 122 as a predetermined event occurs. In addition, the electronic device may store the copied data and metadata as second data D2 and second metadata M2 for the time of the generated event. The electronic device according to the present embodiment can store both data and metadata in the file system area 122 in the new memory 120 according to a predetermined event.

However, even after the first data D1 and the first meta data M1 are copied 73a and 73b to the file system area 122, the electronic device can not perform the data creating operation or the modification operation 71, New data and metadata can be continuously updated as the first data D1 and the first metadata M1 of the main memory area when a writing operation 72 occurs.

Suppose a system error or defect occurs in an electronic device that executes a process. The electronic device may forcibly terminate the running process as described above with reference to FIG. 6 and rollback and restore the status information regarding the process. The electronic device according to the present embodiment may delete the first data D1 and the first meta data M1 stored in the main memory area 121 (74). The electronic device may also roll back the previous second data D2 and the second metadata M2 stored in the file system area 122 back to the main memory area 121. [ Accordingly, the data and the metadata corresponding to the previous process time are restored together, and the durability of the entire system can be expected to be improved.

FIG. 9 is a diagram for explaining a process of performing computation and checkpointing according to an embodiment of the present disclosure.

Referring to Fig. 9, the process of computing and checkpointing electronic devices of the present disclosure is illustrated. The electronic device performs checkpointing when the calculation of the previous step is completed, and then performs the following calculation process.

The electronic device can perform checkpointing on the main memory area. The electronic device may initialize the checkpoint table and set the memory to be read-only when performing checkpointing on the main memory area. Since the memory is set to read-only, no data can be written to pages in the main memory area. The checkpoint table is a table included in the main memory area, and plays a role of writing page information in which data is changed. In this embodiment, the data unit is assumed to be a page, but the data unit can be variously set as bytes, words, 2 bytes, 2 words, and so on.

The electronic device may perform the operations associated with page 10 (80). During operation, the data on page 10 may be updated. However, since the attribute of the memory is read-only, the data of page 10 (80) is not changed. The electronic device writes the page information requiring updating in the checkpoint table, and generates a page in which the page 10 is copied. In Fig. 9, page 14 (80) is a page on which the previous page 10 is copied. Then, the electronic device writes the changed page information in the checkpoint table. As shown in FIG. 9, the checkpoint table 86 includes information that the pages 10 to 14 have been changed.

The electronic device changes the attribute of the memory to write. The electronic device updates page 10. That is, the electronic device has updated page 10 (80a) and copied page 14 (81). The electronic device can update the data through the above-described process. Further, since the electronic device includes copied page information and data, if an error occurs, the data can be easily rolled back using the copied page information. When the operation of the main memory area is completed, the electronic device performs a check pointing process on the main memory area again.

On the other hand, as described above, the electronic device must match the data in the main memory area and the file system area. The file system area includes a buffer and a file storage. If the data in the main memory area has changed, the electronic device writes the changed block 92 into the buffer in the file system area. When the operation and the checkpointing process to the main memory area are performed, the electronic device can have the updated data block 92 only in the buffer while keeping the data block 91 of the file store. If an error occurs in the operation and main memory areas, the electronic device may discard the updated data block 92 stored in the buffer.

When the checkpointing process for the main memory area is completed, the electronic device performs a checkpointing process on the file system area. When performing a checkpointing process on the file system area, the electronic device writes the data block 92 stored in the buffer to the data block 91 of the file store. Even if an error occurs during the checkpointing process for the file system area, since the checkpointing process for the main memory area has already been completed, the electronic device can write the data block 92 stored in the buffer to the file storage. That is, the electronic device can roll forward.

That is, when performing an operation, the electronic device can write the information of the page to the checkpoint table when data update of the original page is requested. The electronic device then generates a copy page that has copied the page and can write the information of the copy page to the checkpoint table. Then, the electronic device can update the data of the original page by changing the attribute of the memory to write. The electronic device may also roll back the original page based on the information written to the checkpoint table if an error occurs when performing an operation or performing checkpointing on the main memory area.

On the other hand, the electronic device may include a file system area, and the file system area may include a buffer and a file storage. The file storage unit may include a block of original data. The electronic device may store the updated data block in a buffer when the data block is updated. The electronic device may then write updated data blocks to the file store when performing checkpointing on the file system area. As described above, the electronic device can perform checkpointing on the main memory area, and then perform checkpointing on the file system area.

Various embodiments have been described so far to maintain data consistency and consistency. A flow chart of the electronic device control method will be described below.

10 is a flowchart of an electronic device control method according to an embodiment of the present disclosure.

Referring to FIG. 10, the electronic device performs check pointing (S1010). The new memory of the electronic device includes a main memory area and a file system area. The electronic device may perform checkpointing on the main memory area and then perform checkpointing on the file system area. The main memory area may include a checkpoint table that writes page information whose data has changed. Then, the electronic device initializes the checkpoint table when performing checkpointing on the main memory area, and can set the page attribute of the memory to be read-only.

The electronic device performs an operation (S1020). The electronic device writes the original page information to the checkpoint table when data update of the original page is requested. Then, the electronic device generates a copy page and also writes copy page information to the checkpoint table. The electronic device can change the memory property to light and update the data on the original page.

11 is a flow diagram illustrating the operation of an electronic device that stores information regarding a particular process in accordance with an event in accordance with an embodiment of the present disclosure;

Referring to Fig. 11, a method of performing checkpointing of an electronic device operating to enhance the durability of the system is shown. The electronic device may determine whether an event designated to perform checkpointing occurs (S1110). Check-pointing may indicate an operation of the electronic device storing a value corresponding to a particular point in time of the changed data and metadata stored in the main memory area for a running process. The electronic device can perform checkpointing based on various events according to preset conditions.

According to the determination result, the electronic device can store information on the first process at the time when the event is generated in the new memory (S1120). More specifically, the electronic device may store an address in the new memory associated with the first process at the time the event occurred. The electronic device may also copy the task structure for the first process into the new memory.

Figure 12 is a flow diagram illustrating the operation of an electronic device that stores specific data and metadata in accordance with an event in accordance with an embodiment of the present disclosure.

Referring to FIG. 12, the electronic device performing checkpointing may determine whether an event designated to perform checkpointing occurs (S1210). According to the determination result, the electronic device can store the first data and the first metadata at the time when the event occurs in the file system (S1220). The electronic device can store the first data and the first metadata at the time when the event occurs according to the determination result of the previous step in the file system. In this embodiment, the metadata may represent secondary data for managing data structurally present in the disc. Metadata can represent data that is created or changed as files are created and deleted, directories are created and deleted, and file sizes increase or decrease. In this embodiment, the first data and the first metadata may represent each of data and metadata associated with a process executed at the time when an event occurs.

If a system error occurs in the electronic device, the second data and the second metadata existing in the main memory area may be deleted, and the first data and the first metadata may be rolled back (S1230). The electronic device may perform a rollback using the first data and the first metadata stored in the file system of the new memory in the event of a system error or defect. More specifically, the electronic device may delete the second data and the second metadata that are present in the main memory area when a system error occurs. Thus, the electronic device can roll back and use the previous first data and the first metadata stored in the file system.

The embodiments described above may be implemented in hardware components, software components, and / or a combination of hardware components and software components. For example, the devices, methods, and components described in the embodiments may be implemented within a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array ), A programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute one or more software applications that are executed on an operating system (OS) and an operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For ease of understanding, the processing apparatus may be described as being used singly, but those skilled in the art will recognize that the processing apparatus may have a plurality of processing elements and / As shown in FIG. For example, the processing unit may comprise a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

The software may comprise a computer program, code, instructions, or a combination of one or more of the foregoing, configured to configure the processing device to operate as desired, or independently or in combination (e.g., collectively) processing devices. The software and / or data may be in the form of any type of machine, component, physical device, virtual equipment, computer storage media, or device , Or may be permanently or temporarily embodied in a transmitted signal wave. The software may be distributed over a networked computer system and stored or executed in a distributed manner. The software and data may be stored on one or more non-volatile computer readable recording media.

A non-transitory readable medium is a medium that stores data for a short period of time, such as a register, cache, memory, etc., but semi-permanently stores data and is readable by the apparatus. In particular, the various applications or programs described above may be stored on non-volatile readable media such as CD, DVD, hard disk, Blu-ray disk, USB, memory card, ROM,

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, It will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the present invention.

100: Electronic device
110: Processor 120: Memory

Claims (16)

A nonvolatile new memory including a main memory area and a file system area; And
And a processor for performing checkpointing on the main memory area and the file system area according to a predetermined condition,
Wherein the main memory area comprises:
And a checkpoint table for writing page information in which the data in the main memory area is changed,
The processor comprising:
Initializing the checkpoint table, setting an attribute of a page of the main memory area as read-only, and updating data of a first page of the main memory area when the first page is updated, Writes information of a page to the checkpoint table, generates a second page in which the first page is copied to roll back, writes information of the second page to the checkpoint table, The attribute of the first page is changed to write and maintained,
The second page is deleted at the checkpoint, the checkpoint table is initialized, the first page which was a write attribute is set to be read-only, and the data of the second page is read from the first page Lt; RTI ID = 0.0 > data,
Wherein the file system area comprises:
A memory buffer for a file in a file system, and a file storage unit, and includes new version data for updated file data and metadata during operation of the system, and at the time of checkpointing, a new version of data block and metadata Applies the data to the file, and erases the data in the main memory buffer when an error occurs, thereby rolling back the electronic device. The method according to claim 1,
The processor comprising:
If an error occurs in the execution of the operation, the second page is rolled back to the first page in the main memory area based on the information written in the checkpoint table, or an operation is performed after checkpointing to the memory file buffer in the file system area And rolls back to the latest checkpoint time. The method according to claim 1,
The processor comprising:
And rolls back the second page to the first page based on information written to the checkpoint table when an error occurs in performing checkpointing on the main memory area. The method according to claim 1,
The processor comprising:
Performing checkpointing according to a predetermined condition for the main memory area, initializing the checkpoint table, and setting a property of a page of the main memory area as read-only. The method according to claim 1,
The file storage unit stores,
A pre-stored block of data,
The processor comprising:
Wherein when the data block is updated, the memory buffer stores new data by storing the updated data block in the memory buffer until checkpointing is performed, and the file storage unit stores multi-version data including previous data , An electronic device. 6. The method of claim 5,
The processor comprising:
And flushes to the file storage all of the pages including the updated metadata and data blocks in the memory buffer when performing checkpointing on the file system area. 6. The method of claim 5,
The processor comprising:
And rolls all pages including the updated data block to the file storage if an error occurs in performing checkpointing on the file system area. The method according to claim 1,
Preferably,
A predetermined period, and a condition that a change in the data satisfies a preset number of times. A control method of an electronic device including a non-volatile new memory including a main memory area and a file system area,
A main memory area checkpointing step of initializing a checkpoint table for writing page information in which data in the main memory area is changed, and setting an attribute of a page of the main memory area as read-only;
A file system area checkpointing step of performing checkpointing on the file system area;
And performing an operation,
Wherein performing the operation further comprises:
If the update of data for the first page of the main memory area is requested, writing the information of the first page to the checkpoint table, and if the second page copied the first page for roll back, Writes the information of the second page to the checkpoint table, changes the attribute of the first page to a write until the next checkpoint interval, and maintains the attribute of the second page in the main memory area when the error occurs, Updating the data of the page to the first page,
Wherein the file system area comprises:
A memory buffer for a file in a file system, and a file storage, and includes new version data for updated file data and metadata during operation of the system,
Wherein the main memory area checkpointing step comprises:
Removing the second page, initializing a checkpoint table, setting the first page that was a write attribute to be read only,
The file system area checkpointing step comprises:
Applying a new version of the data block and metadata in the memory buffer to the file and deleting data in the main memory buffer when an error occurs to roll back. 10. The method of claim 9,
Wherein performing the operation further comprises:
If an error occurs in the execution of the operation, the second page is rolled back to the first page in the main memory area based on the information written in the checkpoint table, or an operation is performed after checkpointing to the memory file buffer in the file system area And rolling back to the latest checkpoint time. 10. The method of claim 9,
Wherein the main memory area checkpointing step comprises:
And rolls back the second page to the first page based on the information written in the checkpoint table when an error occurs in performing checkpointing on the main memory area. 10. The method of claim 9,
Wherein the main memory area checkpointing step comprises:
Performing checkpointing according to a predetermined condition for the main memory area, initializing the checkpoint table, and setting a property of a page of the main memory area as read-only. 10. The method of claim 9,
The file system area checkpointing step comprises:
When the data block previously stored in the file storage unit is updated, the memory buffer stores the updated data block in the memory buffer until the checkpointing is performed, and the file storage unit stores new data, including the previous data, Version of the data. 14. The method of claim 13,
The file system area checkpointing step comprises:
And flushes to the file storage all of the pages including the updated metadata and data blocks in the memory buffer when performing checkpointing on the file system area. 14. The method of claim 13,
The file system area checkpointing step comprises:
And rolls all pages including the updated data block to the file storage when an error occurs in performing checkpointing on the file system area. 10. The method of claim 9,
The file system area checkpointing step comprises:
And flushes the data block of the memory buffer to the file storage after the checkpointing of the first page and the second page of the main memory area is completed.

Images