JP7205090B2 - Control device, control program, and control method - Google Patents

Description

The present invention relates to a control device, control program, and control method.

2. Description of the Related Art In a storage control device that controls a storage device, volatile memory may be backed up to nonvolatile memory when the external power supply is lost (power failure) during operation. Examples of nonvolatile memories include semiconductor memory devices such as SSDs (Solid State Drives).

In backup, data stored in the volatile memory, such as data used for operation of the storage control device, is written into the backup area of the nonvolatile memory using a backup power source such as a battery. As a result, after the external power is turned on again, the storage control device can resume the operation before the power failure by reading the backup data from the backup area of the nonvolatile memory to the volatile memory.

A non-volatile memory such as an SSD has a plurality of blocks as data storage areas. Each block consists of multiple pages.

In nonvolatile memory, when data is repeatedly written and erased to a page or block, the storage area of the nonvolatile memory becomes a state in which valid data is stored discretely in multiple blocks, and all pages are filled. The number of free blocks that are free may decrease.

A non-volatile memory controller may perform an autonomous operation called a garbage collection (GC) operation to create an empty block when the number of empty blocks falls below a certain amount (a fixed number). In GC, the controller collects valid data scattered on multiple blocks and writes them into one free block, erases multiple blocks that are the storage sources of these valid data, and creates multiple free blocks. Repeat the process of creating

JP 2013-200726 A JP 2016-192025 A JP 2013-61799 A

Here, in the backup at power failure, data in the volatile memory is sequentially written in block units to the storage area of the nonvolatile memory. At this time, if there is a shortage of empty blocks that serve as backup write destinations in the storage area, the nonvolatile memory may autonomously execute GC.

Backup during a power failure is performed using a standby power supply, so the time during which backup can be executed is limited according to the power capacity of the standby power supply.

However, if GC occurs during backup during a power failure, the average write speed until the backup is completed decreases, so the backup execution time may increase. In this case, the backup of the memory is not completed within the limited time, and it may be difficult to resume the operation before the power failure occurred after the external power supply of the storage control device is turned on again.

Also, if the backup is to be completed normally even when GC occurs during the backup, the degree of freedom in designing the storage control device may decrease in terms of performance, cost, device scale, and the like.

It should be noted that the above-described inconvenience is not limited to storage control devices, but can also occur in various control devices, such as information processing devices that back up volatile memory to nonvolatile memory such as an SSD when a power failure occurs in the device. can occur as well.

In one aspect, an object of the present invention is to suppress an increase in backup execution time for a nonvolatile memory.

In one aspect, the controller may include a volatile memory, a writer, a determiner, and a transmitter. The writing unit may write dummy data to a backup area provided in the nonvolatile memory at a predetermined timing during operation. The backup area may be a write destination of a backup of data in the volatile memory that is executed in the event of a power failure. The determination unit is a ratio obtained from a controller of the nonvolatile memory, and the control device in the nonvolatile memory determines the ratio of the entire storage area of the nonvolatile memory after the dummy data is written. the volatile memory based on the ratio of the free space including an area not recognized as a storage area of the memory and an empty area in which data is not written among the areas recognized as the storage area of the nonvolatile memory by the control device ; It may be determined whether or not the free space of the nonvolatile memory is insufficient when the backup of is executed. When the determining unit determines that the free space of the nonvolatile memory is insufficient, the transmitting unit causes the controller to increase the free space of the nonvolatile memory. the increase processing for collecting the data discretely written in the memory and writing it in the free space in units of blocks, and releasing the plurality of blocks that are the storage sources of the discretely written data to increase the free space. may send an execution signal to cause the

In one aspect, it is possible to suppress an increase in backup execution time for the nonvolatile memory.

It is a figure explaining an example of operation|movement of GC. 1 is a block diagram showing a functional configuration example of a storage device as one example of an embodiment; FIG. 3 is a diagram showing a data configuration example of a storage unit shown in FIG. 2; FIG. 3 is a diagram showing an example of the state of physical storage blocks in the storage unit shown in FIG. 2; FIG. FIG. 4 is a diagram illustrating an example of GC execution conditions according to an embodiment; FIG. 10 is a diagram showing an example of change in free space remaining capacity due to writing of dummy data; 7 is a flowchart showing an operation example of GC execution determination processing by a CM (Controller Module) according to one embodiment; FIG. 8 is a flowchart showing an operation example of the GC execution process shown in FIG. 7; FIG. 6 is a flow chart showing an operation example of post-power failure detection processing by CM according to one embodiment. 7 is a flowchart showing an operation example of post-power-recovery processing by a CM according to one embodiment; 2 is a block diagram showing a hardware configuration example of a storage device according to one embodiment; FIG.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the embodiments described below are merely examples, and are not intended to exclude various modifications and application of techniques not explicitly described below. For example, this embodiment can be modified in various ways without departing from the spirit of the embodiment. In the drawings used in the following embodiments, parts with the same reference numerals represent the same or similar parts unless otherwise specified.

[1] One Embodiment [1-1] About One Embodiment In the above-described storage control device, the capacity of the volatile memory to be backed up is increasing due to the increase in capacity of DRAM (Dynamic Random Access Memory). On the other hand, the power capacity of backup power sources such as batteries remains constant. Therefore, in order to cope with the expansion of the capacity of the backup target without changing the housing size of the storage control device, the write speed of the non-volatile memory such as the SSD must be improved.

An example of an SSD used as a memory backup destination by a storage control device is an SSD using a NAND flash memory. Since the NAND flash memory does not support data overwrite operation, the controller erases the existing data in advance when rewriting the data. However, erasing data in a NAND flash memory is limited to execution in units of blocks each composed of a plurality of pages.

(Processing 1)
Therefore, when rewriting one page, the controller executes REMW (Read-Erase-Modify-Write). In REMW, the controller reads all pages in the block containing the page to be written to cache in non-volatile memory and erases the block of storage. Then, the controller merges the data to be written into the block data on the cache, and writes the block data to the erased block in the storage area.

(Processing 2)
In the REMW operation in process 1, data is read, erased, changed, and written in block units, which causes a decrease in data writing speed. Therefore, the controller performs block alternation processing. For example, when writing data, if there is an erased empty block, the controller omits erasing of the block in REMW and writes the changed block data to the empty block so that the waiting time for erasing is reduced. to reduce the time required for It should be noted that the controller erases the blocks for which erasing is omitted at a predetermined timing, and uses the blocks as empty blocks for the next writing. In the following description, empty blocks may also be referred to as spare blocks or free blocks.

(Processing 3)
If the data writing and erasing in process 1 and process 2 are repeated, the storage area of the SSD will be in a state where valid data is stored discretely in multiple blocks, and the number of spare blocks may decrease. . Therefore, the controller of the SSD performs the above-described GC operation when the number of empty blocks becomes equal to or less than a certain amount (fixed number).

FIG. 1 is a diagram illustrating an example of GC operation. As shown in (1) of FIG. 1, the storage area (NAND flash memory) is in a state in which the writing from the host device is completely written. In (1) of FIG. 1, each frame indicates a page, white frames are free pages to which data can be written, and hatched frames are pages in which data is written. A set of 8×5 frames indicates a block.

When the remaining free space of the SSD becomes, for example, "Low_Threshold" (threshold) or less due to a write from the host device, the SSD controller performs GC until it reaches a threshold (for example, "High_Threshold") at which GC becomes unnecessary. 1 shows whether or not GC should be performed according to the remaining capacity of the free space of the SSD and its threshold value.For example, the remaining capacity of the free space is GC led by the SSD If the range is "Device Initiated GC", the SSD performs GC so that the range is "No GC" where GC is unnecessary or the range is between "High_Threshold" and "Low_Threshold". Note that GC may or may not be performed in the range between "High_Threshold" and "Low_Threshold".

As shown in (2) of FIG. 1, the SSD actively performs GC when the free space remaining capacity becomes equal to or less than the threshold during host I/O. For example, the SSD writes the pages in the block written incompletely to other free blocks without gaps (see (2-1) in FIG. 1), and erases the block written incompletely into a spare ( free) block (see (2-2) in FIG. 1).

However, since such GC processing competes with host I/O processing, the host I/O has to wait. For example, when GC is executed during host I/O processing, the response (response time) may be about 1.5 to 8 times the response when GC is not executed.

By the way, in one embodiment, a controller module (CM) will be described as an example of a storage control device. The CM may be, for example, a RAID controller that controls storage devices constituting a RAID (Redundant Array of Inexpensive Disks) system.

Also, an SSD may be an example of a backup device (BUD; Back Up Device), and has a memory backup area, a storage area for FW (Firmware) that implements CM operations, and a storage area for system operation logs. area.

For example, the CM irregularly writes logs to a random area in the operation log storage area of the BUD.

In this case, in the BUD, depletion of spare blocks and block replacement processing occur irregularly (processing 1 and processing 2 occur). If a power outage occurs in the storage device during the writing of the log, a large amount of data is written continuously in the backup write operation. Write speed slows down.

Also, when a power failure occurs during operation of the CM, the CM performs sequential LBA (Logical Block Address) sequential writing to the backup area of the BUD as memory backup writing in the event of a power failure.

In this case, inside the BUD, the backup data is concentrated in the page block and written without any gaps. When backup data is erased, unnecessary free blocks of GC are generated. However, since the BUD holds the backup data in a written state (valid data) unless instructed to do so, the blocks in which the backup data are written are not counted as free blocks.

Note that it is conceivable that the CM takes the lead in performing the GC operation of the BUD so as not to cause a shortage of free blocks during backup writing during a power failure, thereby avoiding write performance deterioration. However, for the reasons (i) and (ii) below, the execution condition for the CM to take the lead in executing the GC operation is not unique.

(i) It is difficult to specify the number of BUD free blocks used by backup. For example, the free block capacity (free space) used for memory backup is not uniquely determined from the memory size. Also, when the backup area is overwritten a number of times, written blocks are erased, but the CM does not calculate these blocks as free blocks. Furthermore, since block and page sizes of general SSD devices are often not disclosed, it is difficult for the CM to calculate the number of used blocks from the history of memory sizes that have been backed up.

(ii) It is difficult to determine the number of free blocks present in the BUD when the backup is performed. For example, the consumption status of BUD blocks and pages differs depending on the cumulative state of log writes. Therefore, the number of free blocks in the BUD at the time the backup is executed is indefinite.

In view of the above points, in one embodiment, a technique for suppressing an increase in backup execution time will be described. For example, in one embodiment, in order to prevent the occurrence of a GC operation based on self-determination of the BUD due to the lack of spare blocks in the backup write destination when performing backup, the CM performs GC of the BUD to prevent exhaustion of spare blocks. Take control.

Here, it is assumed that the method according to one embodiment uses SI (Storage Intelligence). SI is a general term for technology that allows a control device such as CM to control the timing of executing internal processing that affects the speed performance of an SSD, such as GC processing, and is being standardized.

A representative function of the SI is, for example, an instruction function for activating GC processing from the control device side outside the SSD. The control device can activate the autonomous GC of the SSD by issuing the command to the SSD at the timing of the control device side. In addition, the instruction can also specify the operating time of GC and the capacity (number) of spare blocks to be created by GC.

[1-2] Configuration Example of Storage Apparatus Next, a functional configuration example of the storage apparatus 1 according to one embodiment will be described with reference to FIG. The storage device 1 is an example of an information processing device, and may include a CM 2 and a storage 3 as illustrated in FIG.

The storage 3 has a storage area provided by the storage device 1 to a host device or the like (not shown). For example, the storage 3 may have multiple disks 3a. For example, the storage 3 may configure a disk array such as RAID with a plurality of disks 3a.

CM2 is an example of a storage controller or controller. The CM 2 may, for example, control the storage device 1 and perform various controls for providing storage areas of the storage 3 to a host device (not shown). In the following description, processing related to access to the storage 3 executed by the CM 2 based on an access request to the storage 3 issued from the host device to the storage device 1 is referred to as "host I/O". sometimes.

As shown in FIG. 2, the CM 2 may exemplarily include a holding unit 10, a memory 20, a BUD 30, a backup processing unit 41, a dummy data writing unit 42, and a GC execution control unit 43.

The holding unit 10 is an example of a memory unit that stores various information used for controlling the CM 2, and may store control information 11, for example. The control information 11 may include, for example, various parameters such as threshold values (TH1, TH2) that are used in processing of the dummy data writing unit 42 and the GC execution control unit 43, which will be described later.

The memory 20 is an example of a volatile memory that stores various information used for the operation of CM2. In one embodiment, the memory 20 is a backup target (backup source) that is executed when a power failure occurs in the storage device 1 (CM2).

The memory 20 may store operation restart information used for restarting the operation of the storage device 1 (CM2), for example, when the storage device 1 (CM2) is restarted after a power failure. The operation restart information includes, for example, control information of the process that the CM 2 was executing immediately before the power failure occurred, cache information related to host I/O, and the like.

The BUD 30 is an example of non-volatile memory that stores various information used for the CM 2 operation. In one embodiment, the BUD 30 is a storage (saving) destination (backup destination) for backup executed when a power failure occurs in the storage device 1 (CM2).

The BUD 30 may have a control section 31 and a storage section 32 . The control unit 31 is an example of a controller for the BUD 30, and controls the storage unit 32 to perform the above-described (processing 1) to (processing 3), for example, control of REMW, control of block replacement processing, and control of GC processing. control, etc. Control of GC processing may be executed under the leadership of the BUD 30 according to the remaining capacity of the free space in the storage unit 32 .

The storage unit 32 is a storage area of the BUD 30, and has a FW storage area 32a, a log meta write area 32b, a backup area 32c, an empty area 32d, and an invisible area 32e as logical blocks, as illustrated in FIG. You can Each of these areas 32a to 32e may be represented as a logical address range by consecutive logical addresses.

The FW storage area 32a is a storage area for storing FW that implements the operation of the CM 2, and the log meta writing area 32b is a storage area for storing system operation logs and the like. The backup area 32c is a storage area for writing a backup of data in the memory 20 that is executed using the standby power supply of the CM2 in the event of a power failure of the CM2. The invisible area 32e is an area that is not recognized by the CM 2 as the capacity of the BUD 30, and may be used by the control unit 31 as a free block together with the empty area 32d in which data has not yet been written.

Next, with reference to FIG. 4, an example of the state of the physical block (physical storage block) corresponding to each of the logical addresses of the storage unit 32, in other words, the areas 32a to 32e shown in FIG. 3 will be described.

FIG. 4 is a diagram showing an example of the state of physical storage blocks. In the example of FIG. 4, for convenience, it is assumed that one frame is a page and a set of four pages is a block. As for the state of the pages, white pages have been erased, shaded pages have valid data and are valid for storage, and hatched pages have invalid data (waiting for erasure) and invalid storage. , respectively.

Also, in the example of FIG. 4, the block enclosed by (a) is used as the FW storage area 32a, and the block enclosed by (b) is used as the log meta writing area 32b. Furthermore, the blocks enclosed by (c) are used as the backup area 32c, and the blocks enclosed by (d) are used as the free area 32d. These areas are visible areas calculated as the storage capacity of the storage unit 32 . On the other hand, a block enclosed by (e) indicates an invisible area 32e that is not calculated as the storage capacity of the storage unit 32. FIG.

In the example of FIG. 4, a block with all pages erased is counted as a free block.

As exemplified in FIG. 4, in order to reduce deterioration of the I/O performance of the BUD 30, the control unit 31 processes GC, block alternation processing, erasure, etc. in the background. In this process, the logical addresses and the physical blocks shown in FIG. 4 are converted sequentially, so the logical-physical allocation in the visible area and the invisible area changes over time. Therefore, as shown in FIG. 4, logical addresses may not be assigned to consecutive physical blocks.

When the CM 2 detects loss of the power supply (external power supply) that supplies power to the CM 2 (detects a power failure), the backup processing unit 41 uses a backup power supply such as a battery to restore the backup area 32c of the BUD 30. Then, the memory 20 is backed up. The data to be backed up may be, for example, valid data stored by the memory 20 when a power failure is detected.

Note that when the backup is completed, the CM 2 stores an operation stop factor (power-off factor) such as "power failure" in the log meta write area 32b of the BUD 30 or other non-volatile memory, etc., and stops the operation. you can When external power is restored to the CM 2 and external power is restored, the backup processing unit 41 refers to the cause of the previous operation stoppage. can be written back to As a result, the CM 2 can resume system operation after restoring the state of the memory 20 when the power failure was detected.

In this way, the backup processing unit 41 is an example of a backup unit that backs up data in the memory 20 to the backup area 32c using the standby power supply when a power failure occurs in CM2.

Here, CM2 periodically writes the operation log in CM2 to the log meta write area 32b of the BUD30. For this reason, due to the writing of the operation log, etc., the storage unit 32 may be in a state in which the data is written to the memory unit 32 as shown in FIG.

In the state illustrated in FIG. 4, for example, while the backup processing unit 41 is performing backup processing due to a power failure, the remaining capacity of the free space in the storage unit 32 becomes equal to or less than "Low_Threshold" (threshold). Suppose (see FIG. 5). In this case, as described above, in the BUD 30, the GC as internal control by the control unit 31 can be executed, so the performance of writing to the BUD 30 in the backup process is degraded.

In one embodiment, the following method is used to suppress the deterioration of write performance by suppressing the occurrence of spontaneous GC (physical GC) by the BUD 30 during backup processing.

The dummy data writing unit 42 periodically issues an instruction to write dummy data to the backup area 32c of the storage unit 32 to the BUD 30 during operation of the CM 2, thereby writing dummy data to the backup area 32c. do The state in which CM2 is operating may mean, for example, a state in which no power failure has occurred and power is being supplied to CM2 normally from an external power source, in other words, the state in which CM2 is operating normally.

For example, dummy data writing unit 42 may perform the following dummy data writing process. The dummy data write process includes, for example, a write process of writing a certain amount of dummy data to the backup area 32c, a determination process of determining whether or not the dummy write end condition is satisfied based on the remaining capacity of the free space of the BUD 30, may contain

Note that the dummy data may be any data, such as actual data in the memory 20, random data, all "0" or "1", and the like. Also, the fixed capacity of dummy data to be written may be, for example, a size of one block to several blocks or several tens of blocks.

For example, the following (I) and (II) are given as conditions for terminating the dummy data write process. The dummy data writing unit 42 may repeatedly execute a set of writing processing and determination processing until it is determined that at least one of these end conditions is satisfied. Incidentally, when it is determined that the termination condition is satisfied, the dummy data writing unit 42 may terminate (stop) the dummy data writing process.

(I) The total amount of written dummy data has reached the capacity of valid data in the memory 20 to be backed up.

If the termination condition (I) above is satisfied, the dummy data writing unit 42 writes dummy data of a size equivalent to (or larger than) the current capacity of valid data in the memory 20 to the backup area 32c, for example. It can be said that the data was written (dummy backup). As a result, it is possible to create a pseudo state of use of the free (backup) blocks of the BUD 30 due to backup processing executed at the time of a power failure, and to predict the remaining capacity of the free blocks.

(II) Free space (or remaining capacity of free space) is saturated.

The above (II) free space saturation will be described below. Free space may mean the total capacity of free blocks, and in the following description, free space (or its remaining capacity) may be referred to as free block (or its remaining capacity) for convenience. FIG. 5 is a diagram illustrating an example of GC execution conditions according to an embodiment.

In one embodiment, the GC execution conditions defined by SI are employed. In the GC execution conditions, for example, as shown in FIG. 5, ranges of "No GC", "Host Initiated GC", and "Device Initiated GC" are defined according to the remaining capacity of the free space in the storage unit 32. .

"No GC" is a state range in which the remaining capacity is greater than "High_Threshold", and GC is unnecessary in this case.

"Host Initiated GC" is a range in which the remaining capacity is greater than "Low_Threshold" and equal to or less than "High_Threshold". you can In addition, in "Host Initiated GC", CM2 (GC execution control unit 43) may perform control including determination of necessity of execution of GC.

"Device Initiated GC" is a range in which the remaining capacity is equal to or less than "Low_Threshold", and in this case, the BUD 30, which is an example of a device, may take the lead in performing GC.

FIG. 6 is a diagram showing an example of a change in the free space remaining capacity due to the writing of dummy data. In the example of FIG. 6, the horizontal axis indicates the number of dummy data write executions (or execution time), in other words, the total dummy data write capacity (total capacity).

The control unit 31 of the BUD 30 writes dummy data to the backup area 32c in response to an instruction from the dummy data writing unit 42, and backs up page combination, GC, block replacement, etc. to the invisible area 32e. Do it on the ground. Thus, in writing dummy data, erasing and writing of blocks are repeated.

Therefore, in the storage unit 32, the free space is not exhausted (the remaining capacity does not become "0"), and at least the free space for the invisible region 32e remains. Therefore, when the writing process and the determination process by the dummy data writing unit 42 are repeatedly executed, in the storage unit 32, the number of free blocks is equal to the free space of the storage unit 32 other than the dummy data and the free space of the backup area 32c. will saturate the total capacity of

If the termination condition (II) above is satisfied, even if the set of write processing and determination processing is repeatedly executed, it is considered that the number of free blocks will not decrease significantly any more. Therefore, regardless of whether or not the termination condition (I) is satisfied, the dummy data writing unit 42 stops the dummy data writing process when the termination condition (II) is satisfied. good.

Note that the dummy data writing unit 42 determines, for example, the amount of change in the remaining capacity of the free block due to the writing process (difference between the remaining capacity due to the current writing and the remaining capacity due to the previous writing) in the determination of (II) above. becomes "0", it may be determined that the remaining capacity is saturated. Note that, in another example, the dummy data writing unit 42 may determine that the remaining capacity is saturated when the amount of change is equal to or less than the saturation threshold. As an example, the saturation threshold may be a value less than several percent.

Further, when a power failure occurs, the dummy data writing unit 42 may stop writing dummy data prior to the backup processing executed by the backup processing unit 41 . As a result, it is possible to prevent the dummy data writing unit 42 from executing the dummy backup during the backup process at the time of power failure.

As described above, the dummy data writing unit 42 is an example of a writing unit that writes dummy data to the backup area 32c provided in the BUD 30 at a predetermined timing during operation. This backup area 32c is a backup area to which backup of data in the volatile memory is written when a power failure occurs.

The GC execution control unit 43 performs GC execution control led by the CM 2 based on the execution result of the dummy data writing process by the dummy data writing unit 42 .

For example, the GC execution control unit 43 determines whether or not the GC operation will occur based on the device's own judgment due to the lack of free blocks in the BUD 30 during the backup process at the time of a power failure, based on the free space when the dummy data write process is completed. You can judge whether

As an example, the GC execution control unit 43 determines that the remaining capacity of the free space after completion of the dummy data write process is equal to or lower than the "High_Threshold" illustrated in FIG. 5 (range of "Host Initiated GC" or "Device Initiated GC"). It may be determined whether or not

When the remaining capacity is equal to or less than "High_Threshold", the GC execution control unit 43 determines (predicts) that the remaining capacity of the free space is in the range of "Host Initiated GC" or "Device Initiated GC" during backup processing at power failure. and may be controlled to perform GC.

As an example, the GC execution control unit 43 may perform the following GC execution control processing on the BUD 30 until the free space remaining capacity becomes greater than the "High_Threshold" (becomes the range of "No GC"). . GC execution control processing includes, for example, processing for transmitting a GC execution instruction specifying a GC execution time (for example, a fixed time) to the BUD 30, and whether or not the remaining free space of the BUD 30 exceeds the "High_Threshold" after GC is completed. may include determination processing for determining the

For example, in the GC execution control process, the GC execution control unit 43 repeats a set of transmission processing and determination processing until the remaining capacity (ratio) of the free space of the BUD 30 exceeds a predetermined ratio, for example, "High_Threshold". can be executed.

In this way, if the free block remaining capacity is less than a predetermined threshold when the dummy data write process is completed, it is predicted that the BUD 30 will initiate GC in the BUD 30 during the backup process (write). be. Therefore, the GC execution control unit 43 instructs the BUD 30 to forcibly execute GC.

As described above, the dummy data writing unit 42 and the GC execution control unit 43 can predict the number of free blocks (total capacity) that will be insufficient during the backup process at the time of a power failure.

In the above description, it is assumed that "High_Threshold" is used as the predetermined threshold in the GC execution control process by the GC execution control unit 43, but it is not limited to this. For example, the GC execution control unit 43 may use "Low_Threshold" (for example, the range of "Device Initiated GC") or a value between "Low_Threshold" and "High_Threshold" as the predetermined threshold.

In addition, the GC execution control unit 43 may perform erasing processing for erasing blocks written in the backup area 32c by dummy data writing processing (dummy backup) by the dummy data writing unit 42 . The deletion process may be executed before the GC execution control process.

Furthermore, when a power failure occurs, the GC execution control unit 43 transmits a GC inhibition instruction (stop instruction) to the BUD 30 to inhibit the GC operation prior to the backup processing executed by the backup processing unit 41. You can Further, the GC execution control unit 43 may transmit an instruction (control signal) to make the storage area in which the dummy data is written in the backup area 32c an empty area. By these instructions, it is possible to suppress the execution of GC according to the instruction from the GC execution control unit 43 during the backup process at the time of power failure, and to increase the free space, so that the writing speed of the backup process can be improved. can be planned.

As described above, the GC execution control unit 43 is an example of a determination unit that determines whether or not the free space of the BUD 30 is insufficient when the memory 20 is backed up based on the result of writing dummy data.

Also, the GC execution control unit 43 is an example of a transmission unit that transmits an execution signal for executing GC to the control unit 31 of the BUD 30 when it is determined that the BUD 30 has insufficient free space. Note that in one embodiment, GC is an example of an increase process for increasing the free space of the BUD 30 .

As described above, according to the method according to one embodiment, the CM 2 that controls the backup, not the control unit 31 of the BUD 30 that is the backup destination, determines whether or not to perform GC in the BUD 30 . Then, when the CM 2 determines to perform GC, it controls the control unit 31 to perform GC in advance in order to prevent the control unit 31 from performing GC during a power failure.

As a result, it is possible to avoid performing GC by the autonomous operation of the BUD 30 in the write processing in the BUD 30 by the memory backup processing that is executed when a power failure occurs. can be planned.

[1-3] Operation Example Next, an operation example of the storage apparatus 1 (CM2) configured as described above will be described with reference to FIGS. 7 to 10. FIG.

[1-3-1] GC Implementation Determination Processing First, an operation example of GC implementation determination processing by CM2 will be described with reference to FIG. Note that the GC implementation determination process shown in FIG. 7 may be performed periodically, for example, at a predetermined timing.

As illustrated in FIG. 7, the dummy data writing unit 42 executes dummy data writing processing. For example, the dummy data writing unit 42 acquires the data size "BU_Size" to be backed up in the memory 20 from the memory 20 (step S1), and sets N=0 as the write offset address N (step S2). ).

Next, the dummy data writing unit 42 positively writes dummy data of a certain capacity (for example, size X) to the offset address N of the backup area 32c of the BUD 30 (step S3).

The dummy data writing unit 42 acquires the free space utilization rate FS(N) from the BUD 30 (step S4). The utilization rate FS(N) may be, for example, a value indicating the ratio of the free space to the (total) storage capacity in the storage unit 32 after dummy data is written to the offset address N.

The dummy data writing unit 42, for example, transmits a command for acquiring the storage capacity, the number of free spaces, etc. to the BUD 30, and uses the free space based on the response transmitted from the BUD 30 in response to the command. A rate FS(N) may be calculated. Note that the dummy data writing unit 42 may acquire the utilization rate FS(N) itself from the BUD 30 if it can be calculated in the BUD 30 . The method of obtaining the free space utilization rate is the same in the following description.
The same is true for

Then, the dummy data writing unit 42 determines whether or not the difference (FS(N)-FS(N-1)) in the utilization rate of the free space from immediately before is "0" (or is equal to or less than the saturation threshold), and/or Alternatively, it is determined whether or not the sum of the written dummy data is equal to or greater than "BU_size" (step S5). The total amount of written dummy data may be obtained, for example, by multiplying the dummy data size X by the number of times of writing (for example, a value (N+1) obtained by adding 1 to the offset address N).

As a result of the determination, if it is determined that the difference in usage rate is not "0" (or less than the saturation threshold) and the sum of the written dummy data is not greater than "BU_size" (No in step S5), the process is executed. Move to step S6. In step S6, the dummy data writing unit 42 adds "1" to the offset address N, and the process proceeds to step S3.

On the other hand, if it is determined as a result of determination that the difference in usage rate is "0" (or less than the saturation threshold), or that the sum of the written dummy data is greater than or equal to "BU_size" (Yes in step S5), The dummy data writing process by the dummy data writing unit 42 ends. Then, the process moves to step S7.

After the dummy data writing process is completed, the GC execution control unit 43 performs the erasing process and the GC execution control process. For example, as the erasing process, the GC execution control unit 43 erases an area (block) in which dummy data has been written in, for example, the backup area 32c of the storage unit 32 to make the area free space (step S7).

In addition, the GC execution control unit 43 determines whether or not FS(N) finally obtained in step S4 is equal to or less than the threshold TH1 as a determination of necessity of execution of the GC execution control process (step S8). The threshold TH1 is an example of a predetermined threshold or a first threshold, and may be "High_Threshold" shown in FIG. 5, for example. If it is determined that FS(N) is not equal to or less than the threshold TH1 (No in step S8), the GC execution control unit 43 determines that CM2 does not need to perform compulsory GC, and the process ends.

On the other hand, if it is determined that FS(N) is equal to or less than the threshold TH1 (Yes in step S8), GC execution processing (GC execution control processing) is performed on the BUD 30 (step S9), and the processing ends.

[1-3-2] GC Execution Processing Next, an operation example of the GC execution processing shown in step S9 of FIG. 7 will be described with reference to FIG.

As illustrated in FIG. 8, the GC execution control unit 43 sets a variable n=0 (step S11), and acquires the current free space utilization rate FS from the BUD 30 as an initial value (step S12).

Next, the GC execution control unit 43 issues a GC execution instruction to the BUD 30 (step S13). The execution instruction may specify, for example, an execution period. The control unit 31 may execute GC within the designated execution period.

Next, the GC execution control unit 43 acquires the free space FS(n) from the BUD 30 (step S14), and determines whether or not FS(n) exceeds the threshold TH2 (step S15). The threshold TH2 is an example of a second threshold, and may be "High_Threshold" shown in FIG. 5, for example. Note that the threshold TH2 may be the same value as the threshold TH1 in step S8 of FIG. 7, and in this case, the threshold TH2 is an example of a predetermined threshold.

When it is determined that FS(n) does not exceed the threshold TH2 (No in step S15), the GC execution control unit 43 adds "1" to the variable n (step S16), and the process proceeds to step S13. do.

On the other hand, if it is determined that FS(n) has exceeded the threshold TH2 (Yes in step S15), the possibility that the number of free blocks will be insufficient during the backup process at the time of a power failure is reduced (free space is increased to an amount that does not require GC). increased), the process ends.

In this way, the GC execution control unit 43 designates a specific execution period as the GC execution unit time, and repeatedly executes the GC for that unit time until the remaining capacity of the free space exceeds the threshold TH2. Increase the capacity of space (number of free blocks).

In step S13, the GC execution period designated by the GC execution control unit 43 may be a fixed period, or may be a variable period depending on the current FS, the number of insufficient free blocks, etc. good. For example, the GC execution control unit 43 performs step The execution period may be determined each time S13 arrives.

[1-3-3] Processing after power failure detection Next, an operation example after a power failure of the storage device 1 (CM2) is detected will be described with reference to FIG. It is assumed that the following processing is executed by a backup power source such as a battery of CM2.

As illustrated in FIG. 9, CM2 forcibly terminates the dummy data writing process by the dummy data writing unit 42 when detecting a power failure (step S21). Also, the CM 2 forcibly stops the GC execution instruction to the BUD 30 by the GC execution control unit 43 (step S22). Through steps S21 and S22, execution of the GC execution determination process shown in FIG. 7 (including the GC execution process shown in FIG. 8) is terminated or stopped.

It should be noted that forced termination or forced termination may include forcibly stopping or terminating the processing without waiting for the bureaucracy of the processing when the processing is being executed. Further, forced termination or forced stop may include stopping (cancelling) the scheduled execution of the process when the process is scheduled to be executed.

Also, the GC execution control unit 43 erases the area (block) in which the dummy data has been written in the backup area 32c of the storage unit 32, and makes the area free space (step S23).

Then, the CM 2 executes power failure processing, for example, backup processing of the memory 20 to the BUD 30 by the backup processing unit 41 (step S24), and the processing ends. Note that the CM 2 may, for example, set the fact that the power is cut off due to a "power failure" in the BUD 30, for example, the log meta write area 32b, as a power cut-off factor (power-off factor), and turn off the power (shut down).

[1-3-4] Processing after Power Recovery Next, an operation example after power recovery, that is, when the power supply to the storage apparatus 1 (CM2) is restarted, will be described with reference to FIG.

As illustrated in FIG. 10, when CM2 is powered on due to restoration of power, it starts system boot processing (step S31). In the system startup process, the CM 2 refers to, for example, the log meta writing area 32b of the BUD 30, and determines whether or not the cause of the previous power cut was a power failure (step S32).

If it is determined that the cause of the power cut immediately before is a power failure (Yes in step S32), the backup processing unit 41 restores data from the BUD 30 to the memory 20 (step S33). For example, the backup processing unit 41 may read the backup data stored in the backup area 32 c (effective data in the memory 20 backed up at the time of power failure) and write it to the memory 20 .

Next, the GC execution control unit 43 erases the backup-written area (block) stored in the backup area 32c of the storage unit 32 (step S34).

The GC execution control unit 43 also instructs the BUD 30 to execute GC (step S35). In this way, after completing the processing for system startup, the CM 2 performs GC execution instruction processing and executes GC of all storage areas in the storage unit 32, thereby maximizing free space at the time of system startup. can be ensured.

As described above, the system activation process is completed (step S36), and the process ends.

On the other hand, when it is determined in step S32 that the cause of power cut immediately before is not a power failure (No in step S32), CM2 performs processing according to the cause of power cut (step S37), and the process proceeds to step S36. .

[1-4] Hardware Configuration Example of Storage Apparatus FIG. 11 is a block diagram showing a hardware configuration example of a storage apparatus 1 as an example of an embodiment. As shown in FIG. 11, the storage apparatus 1 illustratively includes a processor 1a, a DRAM 1b, a flash ROM (Read Only Memory) 1c, an SSD 1d, a NWIF (Network Interface) 1e, and a channel IF 1f. Further, the storage device 1 may illustratively include device IFs 1g-1 and 1g-2, a plurality of disks 1h, a battery 1i, an I/O (Input/Output) section 1j, and a reading section 1k.

The processor 1a is an example of an arithmetic processing device that performs various controls and operations. The processor 1a may be communicably connected to each of the blocks 1b to 1k via a bus 1p. As the processor 1a, an integrated circuit (IC) such as a CPU, MPU, DSP, ASIC, PLD (for example, FPGA) may be used. Note that CPU is an abbreviation for Central Processing Unit, MPU is an abbreviation for Micro Processing Unit, and DSP is an abbreviation for Digital Signal Processor. ASIC is an abbreviation for Application Specific Integrated Circuit, PLD is an abbreviation for Programmable Logic Device, and FPGA is an abbreviation for Field Programmable Gate Array.

DRAM 1b is an example of a volatile memory that stores various data and programs. In one embodiment, at least a partial storage area of the DRAM 1b may be used as a cache for temporarily storing various data related to access to the disk 1h from a host device (not shown), for example.

The flash ROM 1c is an example of a nonvolatile memory that stores various data, programs, and the like. For example, the flash ROM 1c may store an OS (Operating System), FW, programs such as applications, and various data. The flash ROM 1c includes various nonvolatile memories such as NAND flash memory, SCM (Storage Class Memory), ROM (Read Only Memory), and the like.

The SSD 1d is an example of nonvolatile memory that stores various data, programs, and the like. The SSD 1d may, for example, have a controller 1da that controls various accesses to the storage area. The controller 1da includes an integrated circuit (IC) including a processor and memory. Processors include various arithmetic processing units. Memory includes various volatile or non-volatile memories.

For example, the SSD 1d may store an OS, FW, programs such as applications, and various data. The SSD 1d includes, for example, an SSD using a NAND flash memory that supports an IF such as SATA, SAS, or NVMe. SATA is an abbreviation for Serial ATA (Advanced Technology Attachment), and SAS is an abbreviation for Serial Attached SCSI (Small Computer System Interface). NVMe is an abbreviation for Non-Volatile Memory Express. As the SSD 1d, for example, a low-latency SSD such as Intel Optane SSD (registered trademark) employing 3D XPoint (registered trademark) technology may be used.

At least one of the flash ROM 1c and the SSD 1d may store a program 1m that implements all or part of various functions of the CM2. For example, the processor 1a of CM2 may implement the function of CM2 by reading out the program 1m as FW stored in the flash ROM 1c, developing it in the DRAM 1b, and executing it. Alternatively, the processor 1a of the CM 2 may read the program 1m as the FW stored in the SSD 1d, develop it in the DRAM 1b or the flash ROM 1c, and execute it, thereby realizing the function of the CM 2. The functions of the CM 2 may include the functions of the backup processing unit 41, the dummy data writing unit 42, and the GC execution control unit 43 described above.

Note that the holding unit 10 shown in FIG. 2 may be realized by, for example, at least one storage area of the DRAM 1b, the flash ROM 1c, and the SSD 1d of the CM2. Also, the memory 20 shown in FIG. 2 may be implemented by, for example, a storage area of the DRAM 1b of the CM2. Furthermore, the BUD 30 shown in FIG. 2 may be implemented by, for example, the storage area of the SSD 1d of CM2.

The NWIF 1e and the channel IF 1f are examples of communication IFs that control connection and communication with LANs (Local Area Networks) and SANs (Storage Area Networks), respectively. These communication IFs may include, for example, an IF such as a LAN card or an IF conforming to Fiber Channel (FC) or the like.

The CM 2 may have a communication IF for controlling connection and communication with the management terminal of the administrator, and may use the communication IF to download the program 1m from a network (not shown).

Device IFs 1g-1 and 1g-2 are examples of communication IFs that control connections and communications with a plurality of disks 1h, respectively. Device IFs 1g-1 and 1g-2 include, for example, IFs such as device adapters and switch devices such as SAS expanders.

A plurality of disks 1h is an example of hardware that stores various data, programs, and the like. For example, the disk 1h may be used as a storage device that provides the storage area of the storage device 1 to a host device or the like (not shown). Examples of the disk 1h include magnetic disk devices such as HDDs (Hard Disk Drives), semiconductor drive devices such as SSDs, and various storage devices such as nonvolatile memories. Examples of nonvolatile memory include flash memory, SCM, ROM, and the like. In one embodiment, the plurality of disks 1h is an example of the plurality of disks 3a forming the storage 3 shown in FIG.

The battery 1i is an example of a standby power source that supplies standby power to the storage device 1 (CM2) when the power source that supplies power to the storage device 1 (CM2) is lost (when a power failure occurs). . The battery 1i may, for example, include control circuitry for detecting loss of power and controlling the supply of reserve power in response to such detection.

The I/O unit 1j may include, for example, one or both of an input unit such as a mouse, keyboard, touch panel, and operation buttons, and an output unit such as a display and printer. The reading unit 1k is an example of a reader that reads data and programs recorded on the recording medium 1n and outputs them to the processor 1a. The reading unit 1k may include a connection terminal or device to which the recording medium 1n can be connected or inserted. Examples of the reading unit 1k include an adapter conforming to USB (Universal Serial Bus), a drive device for accessing a recording disk, and a card reader for accessing flash memory such as an SD card. Note that the program 1m may be stored in the recording medium 1n.

Examples of the recording medium 1n include non-temporary computer-readable recording media such as magnetic/optical discs and flash memories. Examples of magnetic/optical discs include flexible discs, CDs (Compact Discs), DVDs (Digital Versatile Discs), Blu-ray discs, and HVDs (Holographic Versatile Discs). Examples of flash memories include semiconductor memories such as USB memories and SD cards. Examples of CD include CD-ROM, CD-R, CD-RW, and the like. Examples of DVD include DVD-ROM, DVD-RAM, DVD-R, DVD-RW, DVD+R, and DVD+RW.

The hardware configuration of the storage device 1 described above is an example. Therefore, the hardware in the storage device 1 or in the CM 2 may be increased or decreased (for example, addition or deletion of arbitrary blocks), division, integration in arbitrary combinations, addition or omission of buses, etc. may be performed as appropriate.

For example, the disk 1h that configures the storage 3 may be configured as a module separate from the CM 2 or the storage device 1, such as a device enclosure (DE). Further, the blocks 1a to 1k as CM2 may be made redundant (for example, duplicated) in the storage device 1. FIG. If the storage device 1 has a plurality of CMs 2, the access paths between each of these CMs 2 and the storage 3 may also be made redundant. In addition, the storage device 1 may have, for example, a plurality of storages 3 connected in cascade (connected in series).

[2] Others The technique according to the embodiment described above can be modified and changed as follows.

For example, each functional block of CM2 shown in FIG. 2 may be combined in an arbitrary combination or divided.

Also, in one embodiment, the CM2 is used as an example of the control device, but the control device is not limited to this. As the control device, for example, various information processing devices having a nonvolatile memory such as an SSD as a memory backup destination in the event of a power failure may be used.

[3] Supplementary Note The following Supplementary Note will be disclosed with respect to the above embodiment.

(Appendix 1)
volatile memory;
A backup area, which is a write destination for backing up data in the volatile memory that is executed in the event of a power failure, and is written to the backup area provided in the nonvolatile memory at a predetermined timing during operation. a writing unit that writes dummy data;
a determination unit that determines whether or not the free space of the nonvolatile memory is insufficient when the backup of the volatile memory is executed based on the result of writing the dummy data;
When the determination unit determines that the free space of the nonvolatile memory is insufficient, an execution signal is transmitted to the controller of the nonvolatile memory to execute an increase process for increasing the free space of the nonvolatile memory. a transmitter;
control device.

(Appendix 2)
The determination unit acquires a remaining capacity ratio of the free space in the nonvolatile memory from the controller, and determines that the free space in the nonvolatile memory is insufficient when the remaining space ratio is equal to or less than a predetermined threshold. ,
1. The control device according to appendix 1.

(Appendix 3)
The writing unit writes the dummy data into the backup area in units of blocks, and writes the remaining capacity of the free area of the nonvolatile memory each time a block that is part of the dummy data is written into the backup area. is saturated, and if it is determined that the remaining capacity of the free space is saturated, the writing of the dummy data is terminated;
The control device according to appendix 1 or appendix 2.

(Appendix 4)
The data size of the dummy data is equal to or larger than the data size of the data in the volatile memory.
The control device according to any one of Appendices 1 to 3.

(Appendix 5)
5. The method according to any one of appendices 1 to 4, further comprising a backup unit that backs up the data in the volatile memory in the backup area using a standby power supply in the event of a power failure. Control device as described.

(Appendix 6)
When the power failure occurs, before the backup unit executes the backup,
The writing unit uses the standby power supply to stop writing the dummy data to the backup area,
The transmission unit uses the standby power supply to transmit a stop signal for stopping execution of the increase processing to the controller, and the write unit in the backup area writes the dummy data. sending a control signal to make the storage area free;
The control device according to appendix 5.

(Appendix 7)
The backup unit restores the data stored in the backup area of the nonvolatile memory when it is determined that the cause of the previous power cutoff of the control device is a power failure after the power is turned on to the control device. write to volatile memory,
The control device according to appendix 5 or appendix 6.

(Appendix 8)
The execution signal includes the execution time of the increase process,
The transmission unit repeatedly transmits the execution signal to the controller until a remaining capacity ratio of the free area of the nonvolatile memory after the increase processing is executed by the execution signal exceeds a predetermined ratio. do,
The control device according to any one of Appendices 1 to 7.

(Appendix 9)
A backup area, which is a write destination for backing up data in a volatile memory that is executed in the event of a power failure. write data,
determining whether the free space of the nonvolatile memory is insufficient when the backup of the volatile memory is executed based on the result of writing the dummy data;
When it is determined that the free space of the nonvolatile memory is insufficient, an execution signal is sent to the controller of the nonvolatile memory to execute an increase process for increasing the free space of the nonvolatile memory.
A control program that causes a computer to execute a process.

(Appendix 10)
The determination acquires the ratio of the remaining capacity of the free space in the nonvolatile memory from the controller, and determines that the free space of the nonvolatile memory is insufficient when the ratio of the remaining capacity is equal to or less than a predetermined threshold.
The control program according to appendix 9.

(Appendix 11)
The writing
Each time the dummy data is written to the backup area in units of blocks and a block that is part of the dummy data is written to the backup area, whether or not the remaining capacity of the free area of the nonvolatile memory is saturated. to determine
terminating the writing of the dummy data when it is determined that the remaining capacity of the free space is saturated;
The control program according to appendix 9 or appendix 10.

(Appendix 12)
The data size of the dummy data is equal to or larger than the data size of the data in the volatile memory.
The control program according to any one of Appendices 9 to 11.

(Appendix 13)
backing up the data in the volatile memory to the backup area using a standby power supply when the power failure occurs;
13. The control program according to any one of appendices 9 to 12, which causes the computer to execute processing.

(Appendix 14)
If the power failure occurs, before executing the backup,
stopping the writing of the dummy data to the backup area using the standby power supply;
using the standby power supply to transmit a stop signal to stop execution of the increase process to the controller;
transmitting a control signal for making the storage area in the backup area, in which the dummy data is written by the writing unit, an empty area;
14. The control program according to appendix 13, which causes the computer to execute a process.

(Appendix 15)
After powering on the computer, if it is determined that the cause of the previous power-off of the computer was a power failure, the data stored in the backup area of the nonvolatile memory is written to the volatile memory;
The control program according to appendix 13 or appendix 14.

(Appendix 16)
The execution signal includes the execution time of the increase process,
The execution signal is transmitted to the controller until the ratio of the remaining capacity of the free area of the nonvolatile memory after the increase processing is executed by the execution signal exceeds a predetermined ratio. send repeatedly,
The control program according to any one of Appendices 9 to 15.

(Appendix 17)
A backup area, which is a write destination for backing up data in a volatile memory that is executed in the event of a power failure. write data,
determining whether the free space of the nonvolatile memory is insufficient when the backup of the volatile memory is executed based on the result of writing the dummy data;
When it is determined that the free space of the nonvolatile memory is insufficient, an execution signal is sent to the controller of the nonvolatile memory to execute an increase process for increasing the free space of the nonvolatile memory.
control method.

(Appendix 18)
The determination acquires the ratio of the remaining capacity of the free space in the nonvolatile memory from the controller, and determines that the free space of the nonvolatile memory is insufficient when the ratio of the remaining capacity is equal to or less than a predetermined threshold.
The control method according to appendix 17.

(Appendix 19)
The writing
Each time the dummy data is written to the backup area in units of blocks and a block that is part of the dummy data is written to the backup area, whether or not the remaining capacity of the free area of the nonvolatile memory is saturated. to determine
terminating the writing of the dummy data when it is determined that the remaining capacity of the free space is saturated;
The control method according to appendix 17 or appendix 18.

(Appendix 20)
The data size of the dummy data is equal to or larger than the data size of the data in the volatile memory.
The control method according to any one of Appendices 17-19.

1 storage device 2 CM
3 storage 3a disk 10 holding unit 11 control information 20 memory 30 BUD
31 control unit 32 storage unit 41 backup processing unit 42 dummy data writing unit 43 GC execution control unit

Claims (6)

volatile memory;
A backup area, which is a write destination for backing up data in the volatile memory that is executed in the event of a power failure, and is written to the backup area provided in the nonvolatile memory at a predetermined timing during operation. a writing unit that writes dummy data;
A ratio obtained from a controller that the nonvolatile memory has, and the control device in the nonvolatile memory as the storage area of the nonvolatile memory with respect to the entire storage area of the nonvolatile memory after the dummy data is written Backup of the volatile memory is performed based on the ratio of the free space including the unrecognized area and the empty area in which data is not written in the area recognized by the control device as the storage area of the nonvolatile memory. a determination unit that determines whether or not the free space of the nonvolatile memory is insufficient when
When the determining unit determines that the free space of the nonvolatile memory is insufficient, an increasing process for increasing the free space of the nonvolatile memory for the controller , comprising: an execution signal for executing the increasing process of collecting the data written in the above-mentioned manner and writing the data into the free space in units of blocks, releasing the plurality of blocks that are the storage sources of the discretely written data, and increasing the free space. a transmitter that transmits
control device. The determining unit acquires the ratio from the controller, and determines that the free space of the nonvolatile memory is insufficient when the ratio is equal to or less than a predetermined threshold.
A control device according to claim 1 . The data size of the dummy data is equal to or larger than the data size of the data in the volatile memory.
The control device according to claim 1 or 2 . The execution signal includes the execution time of the increase process,
The transmission unit repeatedly transmits the execution signal to the controller until the ratio after the increase processing is executed by the execution signal exceeds a predetermined ratio.
A control device according to any one of claims 1 to 3 . A backup area, which is a write destination for backing up data in a volatile memory that is executed in the event of a power failure. write data,
A ratio obtained from a controller included in the nonvolatile memory, which is recognized as a storage area of the nonvolatile memory by a computer in the nonvolatile memory with respect to the entire storage area of the nonvolatile memory after the dummy data is written. When the backup of the volatile memory is executed based on the ratio of the free space including the free space where no data is written in the area recognized by the computer as the storage area of the non- volatile memory determines whether the free space of the non-volatile memory is insufficient,
When it is determined that the free space of the non-volatile memory is insufficient, increasing processing for increasing the free space of the non-volatile memory for the controller , wherein the data discretely written to the non-volatile memory are collected and written into the free space in units of blocks, and an execution signal is transmitted to execute the increase process of releasing the plurality of blocks that are the storage sources of the discretely written data and increasing the free space .
A control program that causes the computer to execute processing. A backup area, which is a write destination for backing up data in a volatile memory that is executed in the event of a power failure. write data,
A ratio obtained from a controller included in the nonvolatile memory, which is recognized as a storage area of the nonvolatile memory by a computer in the nonvolatile memory with respect to the entire storage area of the nonvolatile memory after the dummy data is written. When the backup of the volatile memory is executed based on the ratio of the free space including the free space where no data is written in the area recognized by the computer as the storage area of the non- volatile memory determines whether the free space of the non-volatile memory is insufficient,
When it is determined that the free space of the non-volatile memory is insufficient, increasing processing for increasing the free space of the non-volatile memory for the controller , wherein the data discretely written to the non-volatile memory are collected and written into the free space in units of blocks, and an execution signal is transmitted to execute the increase process of releasing the plurality of blocks that are the storage sources of the discretely written data and increasing the free space .
A control method in which the computer executes the processing .

Images