JP5630313B2 - Storage control device, storage system, storage control method, and program therefor - Google Patents

Description

  The present invention relates to a storage control apparatus, a storage system, a storage control method, and a program therefor, and more particularly, to a storage apparatus, a storage control apparatus, a storage system, a storage control method, and a program therefor for power saving of the storage system.

  Various related technologies for power saving of storage devices and storage systems are known.

  For example, Patent Document 1 discloses a computer and a method for managing a storage device that calculate a necessary storage capacity based on information on usage type and capacity increase tendency, and can reduce an excess storage capacity. . As described above, Patent Document 1 discloses a power saving technique in a general storage system.

  There is a deduplication storage different from such a general storage system.

  The deduplication storage only updates the management table of duplicate data and does not save the input data in the storage when the input data is already duplicated with data that has been input and stored in the past. In such deduplication storage, when most of the input data overlaps with data stored in the past, the capacity of data actually written to the storage is small. Accordingly, the storage utilization rate tends to be low.

  Patent Documents 2 and 3 describe technologies related to power saving control in such deduplication storage.

  Patent Document 2 discloses a storage system and a power consumption reduction method for the storage system. This storage system has a plurality of logical units including a plurality of disk devices and a power supply control unit. The power supply control unit individually turns on / off each logical unit. Each of the plurality of logical units is assigned to a constantly operating group or a power saving group. This storage system operates to aggregate data having the same hash value into the same logical unit. By doing so, this storage system controls so that the number of logical units belonging to the power saving group to be powered on is reduced when deduplication is executed.

  Further, the power efficient storage using the data deduplication described in Patent Document 3 has a buffer volume in which the power is always turned on and a base volume in which the power is normally turned off. In this storage, the base volume is turned on at a specific timing, and the data stored in the buffer volume is compared with the data stored in the base volume. Based on the comparison result, the storage stores, in the base volume, data that does not overlap with the data stored in the base volume among the data stored in the buffer volume. Thereafter, the storage powers off the base volume.

JP 2010-211743 A JP 2009-059096 A JP 2009-020858 A

  However, the related technology described above has a problem that power saving in the deduplication storage is insufficient.

  The reason why power saving is insufficient is as follows.

  This is because the storage system and the power consumption reduction method of the storage system of Patent Document 2 have a logical unit that is always in operation mode regardless of the deduplication status.

  This is because the power efficient storage using the data deduplication of Patent Document 3 has a buffer volume that is always operating. In other words, this storage always operates a buffer volume with a large capacity that can store all input data from the start of the base volume to the execution of the duplicate determination process regardless of the amount of duplicate data. Because.

  An object of the present invention is to provide a storage control device, a storage system, a storage control method, and a program therefor that can solve the above-described problems.

The storage control device according to the present invention is a non-overlapping data block in which an input data block that is a candidate for data to be stored in the storage device includes data that is different from data included in any stored data block already stored in the storage device Is determined by comparing the hash value of the data included in the input data block and the hash value of the data included in each storage data block, and outputs the determination result of the determination When,
A buffer holding the non-overlapping data block;
A management information storage unit that stores a hash value and a storage destination address in association with each storage data block and each non-overlapping data block;
Based on the determination result, the capacity of the buffer, and the current used capacity of the buffer, prediction information corresponding to prediction of a future occurrence state of the non-duplicated data is calculated, and the calculated prediction information is output. A duplicate data increase prediction unit;
When the storage device has been activated, the non-overlapping data block stored in the buffer is output to the storage device, and the output non-overlapping stored in the management information storage unit A buffer output control unit for updating the storage destination address corresponding to the data block;
A mode control unit that determines an operation mode of the storage device based on the presence / absence of output of the prediction information and the prediction information, and notifies the storage device of the determined operation mode. The non-duplicate data block is stored in the buffer, and the set of the hash value and the storage destination address corresponding to the stored non-duplicate data block is added to the management information storage unit.

In the storage control method of the present invention, a storage control device having a buffer and a management information storage unit
Whether the input data block that is a candidate for data to be stored in the storage device is a non-overlapping data block including data different from the data included in any storage data block already stored in the storage device, Judging by comparing the hash value of the data included in the input data block and the hash value of the data included in each storage data block, and outputting the determination result of the determination,
Storing the non-duplicate data block in the buffer, adding the set of the hash value and the storage destination address corresponding to the stored non-duplicate data block to a management information storage unit;
Based on the determination result, the capacity of the buffer and the current used capacity of the buffer, calculate prediction information corresponding to the prediction of the future occurrence state of the non-duplicated data, and output the calculated prediction information,
When the storage device has been activated, the non-overlapping data block stored in the buffer is output to the storage device, and the output non-overlapping stored in the management information storage unit Update the destination address corresponding to the data block;
The operation mode of the storage device is determined based on whether or not the prediction information is output and the prediction information, and the determined operation mode is notified to the storage device.

The program of the present invention is stored in a computer having a buffer and a management information storage unit.
Whether the input data block that is a candidate for data to be stored in the storage device is a non-overlapping data block including data different from the data included in any storage data block already stored in the storage device, Judging by comparing the hash value of the data included in the input data block and the hash value of the data included in each storage data block, and outputting the determination result of the determination,
Storing the non-duplicate data block in the buffer, adding the set of the hash value and the storage destination address corresponding to the stored non-duplicate data block to a management information storage unit;
Based on the determination result, the capacity of the buffer and the current used capacity of the buffer, calculate prediction information corresponding to the prediction of the future occurrence state of the non-duplicated data, and output the calculated prediction information,
When the storage device has been activated, the non-overlapping data block stored in the buffer is output to the storage device, and the output non-overlapping stored in the management information storage unit Update the destination address corresponding to the data block;
Based on the output of the prediction information and the prediction information, the operation mode of the storage device is determined, and the storage device is notified of the determined operation mode.

  The present invention has an effect that sufficient power saving can be achieved in the deduplication storage.

It is a block diagram which shows the structure of the 1st and 2nd embodiment of this invention. It is a figure which shows the example of the hash value management table in the 1st and 2nd embodiment of this invention. It is a figure which shows the example of the reference source management table in the 1st and 2nd embodiment of this invention. It is a figure which shows the example of the duplication determination log | history in the 1st and 2nd embodiment of this invention. It is an equation for predicting the future occurrence of non-redundant data according to the first and second embodiments of the present invention. It is a flowchart which shows operation | movement of the deduplication part of the 1st and 2nd embodiment of this invention. It is a flowchart which shows operation | movement of the non-redundant data increase prediction part of the 1st and 2nd embodiment of this invention. It is a flowchart which shows operation | movement of the mode control part of the 1st and 2nd embodiment of this invention. It is a block diagram which shows the structure of the 2nd Embodiment of this invention. It is a block diagram which shows the structure of the storage control apparatus which makes a computer perform a predetermined | prescribed process with the program of this invention.

  Next, embodiments of the present invention will be described in detail with reference to the drawings.

[First Embodiment]
FIG. 1 is a block diagram showing the configuration of the first exemplary embodiment of the present invention.

  Referring to FIG. 1, the present embodiment includes a storage control device 100, a storage device 200, and a host system 300.

  The storage control device 100, the storage device 200, and the host system 300 are connected by a network 900. Communication among the storage control device 100, the storage device 200, and the host system 300 is performed using, for example, IP (Internet Protocol). Further, for example, the reading and writing of data with respect to the storage apparatus 200 may take the form of a SAN (Step Storage Area Network).

  The storage control device 100 includes a data dividing unit 110, a duplication determination unit 120, a non-duplication data increase prediction unit 130, a mode control unit 140, a buffer output control unit 150, a management information storage unit 160 and a buffer 170.

  The storage device 200 includes a power supply control unit 210 and one or more disk units (also called data storage units) 220.

  The host system 300 transmits a write command including input data and a write address to be written to the storage apparatus 200.

  In the storage control device 100, the data dividing unit 110 receives a write command from the host system 300, and divides the input data to generate one or more input data blocks (data blocks that are candidates for storage in the storage device 200). . At the same time, the data dividing unit 110 generates a write address (hereinafter referred to as a reference source address) for each generated input data block based on the write address included in the write command. Next, the data division unit 110 outputs a set of the generated input data block and the corresponding reference source address to the duplication determination unit 120. Here, the reference address is a logical address when the host system 300 accesses the storage apparatus 200.

  Note that the size of the generated input data block may be a variable length or a fixed length. In the present embodiment, the description will be made assuming that the size of the input data block is variable. Since the technique for dividing the input data is well known, a description thereof will be omitted.

  The management information storage unit 160 stores a hash value management table 161 as shown in FIG. 2 and a reference source management table 162 as shown in FIG.

  FIG. 2 is a diagram illustrating an example of the hash value management table 161. Referring to FIG. 2, the hash value management table 161 includes hash values corresponding to input data blocks (non-duplicate data blocks described later) held in the buffer 170 and stored data blocks stored in the storage apparatus 200. 163 and a storage destination address 164 are included. The stored data block is an input data block generated by the data dividing unit 110 and is a data block already stored in the storage apparatus 200.

  FIG. 3 is a diagram illustrating an example of the reference source management table 162. Referring to FIG. 3, the reference source management table 162 includes a pair of a reference source address 165 and a storage destination address 164 corresponding to each non-overlapping data block and each stored data block.

  The storage destination address 164 corresponding to the storage data block is a physical address when the storage control device 100 accesses the storage device 200. A storage destination address 164 corresponding to a non-overlapping block is an address on the buffer 170. The reference source address 165 is a logical address when the host system 300 accesses the storage apparatus 200.

  The duplication determination unit 120 determines whether or not the input data block is a non-duplication data block including data different from the data included in the storage data block, and determines the determination result information including the determination result as non-duplication data. Output to the increase prediction unit 130.

  Specifically, the duplication determination unit 120 calculates a hash value (hereinafter referred to as a hash value 663, not shown) of data included in the input data block. Since the technique for calculating the hash value 663 is well known, the description thereof is omitted.

  Subsequently, the duplication determination unit 120 compares the hash value 663 with the hash value 163 of the hash value management table 161. If the hash value 663 does not match any hash value 163, the duplication determination unit 120 determines that the corresponding input data block is a non-duplication data block. On the other hand, when the hash value 663 matches any one of the hash values 163, the duplication determination unit 120 determines that the corresponding input data block is a duplication data block.

  When the duplication determination unit 120 determines that the corresponding input data block is a non-duplication data block, the duplication determination unit 120 writes the non-duplication data block in the buffer 170. At the same time, the duplication determination unit 120 updates the hash value management table 161 and the reference source management table 162. Specifically, the duplication determination unit 120 adds a set of the hash value 163 and the storage destination address 164 of the non-duplication data block to the hash value management table 161. Similarly, the duplication determination unit 120 adds a set of the reference source address 165 and the storage destination address 164 of the non-overlapping data block to the reference source management table 162. In this case, the storage destination address 164 is an address on the buffer 170.

  In addition, when it is determined that the corresponding input data block is not a non-duplicate data block, that is, a duplicate block, the duplicate determination unit 120 discards the duplicate data block. At the same time, the duplication determination unit 120 updates the reference source management table 162. Specifically, the duplication determination unit 120 updates the reference source management table 162 as follows, for example. First, the duplication determination unit 120 refers to the hash value management table 161 and acquires a storage destination address 164 corresponding to the hash value 163 having the same value as the hash value 663 of the input block data. Next, the duplication determination unit 120 adds the reference source address 165 of the input block data and the acquired storage destination address 164 to the reference source management table 162 as a pair.

  Next, the duplication determination unit 120 outputs determination result information including the size of the input data block corresponding to the determination result and the input data throughput. In the present embodiment, the input data throughput is the amount of input data per unit time. For example, the duplication determination unit 120 calculates the input data throughput by adding up the sizes of the input data blocks received in the past one second from the current time.

  The buffer 170 holds non-overlapping data blocks. The capacity (B_max bytes) included in the buffer 170 is determined as follows, for example. It is assumed that the maximum capacity per second of input data received by the storage control device 100 is B_input bytes / second. It is assumed that the startup time required for the storage apparatus 200 to change from the power saving mode state to the normal operation mode state is T_start seconds. In this case, the capacity (B_max bytes) of the buffer 170 is larger than (B_input × T_start) bytes.

  The buffer output control unit 150 outputs the input data block (non-overlapping data block) stored in the buffer 170 to the storage device 200. Subsequently, the buffer output control unit 150 clears the buffer 170. At the same time, the buffer output control unit 150 updates the storage destination address 164 of the hash value management table 161 and the reference source management table 162 from the address on the buffer 170 to the address on the storage device 200.

  The non-redundant data increase prediction unit 130 responds to the future occurrence state of predicted non-duplicate data based on the determination result information output from the duplication determination unit 120, the capacity of the buffer 170, and the current holding amount of the buffer 170. The prediction information to be calculated is calculated. Next, the non-overlapping data increase prediction unit 130 outputs the prediction information. The current holding amount of the buffer 170 is the total size of the input data blocks held in the buffer 170 acquired from the buffer 170 by the non-redundant data increase prediction unit 130.

  For example, the non-overlapping data increase prediction unit 130 calculates prediction information as follows. First, the non-overlapping data increase prediction unit 130 determines all the data sizes of the input data block during a predetermined period (for example, the past 1 millisecond) based on the size of the input data block and the corresponding determination result. A ratio (Y ÷ X) between the total value (X) and the total value (Y) of all data sizes of non-overlapping data blocks is calculated. Note that “during a predetermined period” is also referred to as “conforming to a predetermined condition”. “Conforming to a predetermined condition” may be, for example, “within a predetermined number of ranges from the latest input data block”.

  Next, the non-redundant data increase prediction unit 130 calculates the non-redundant data throughput by multiplying the calculated ratio by the input data throughput. Next, the non-duplicate data increase prediction unit 130 determines the size of the non-duplicate data block held in the buffer 170 based on the calculated non-duplicate data throughput, the capacity of the buffer 170, and the current holding amount of the buffer 170. The time until the total reaches the capacity of the buffer 170 is calculated as prediction information.

  More specifically, the non-overlapping data increase prediction unit 130 calculates prediction information as follows, for example.

  First, the non-redundant data increase prediction unit 130 shows the determination result output by the duplication determination unit 120 and the size of the corresponding input data block in a storage unit (not shown) in the non-redundant data increase prediction unit 130, as shown in FIG. Such duplication determination history 131 is recorded. FIG. 4 is a diagram illustrating an example of the duplication determination history 131. Referring to FIG. 4, the overlap determination history 131 includes a time 132, a size 133, and a determination result 134. The time 132 indicates the time when the determination result information output from the duplication determination unit 120 is received, for example, in units of microseconds and in a cycle of 10 milliseconds. The size 133 indicates the size of the received input data block, for example, in kilobytes. The determination result 134 indicates that the data block is a non-overlapping data block when “1”, and that the data is redundant data when “0”.

  The non-overlapping data increase prediction unit 130 records the data size of the input data block for the past 1 millisecond each time the duplication determination history 131 is recorded (size 133 corresponding to the input data block whose determination result 134 is 0 or 1). And the total value of all the data sizes of non-overlapping data blocks (the size 133 corresponding to the input data block whose determination result 134 is 1). The ratio (Y ÷ X) is calculated. In the example of FIG. 4, when the current time is the time 132 “2601” that is the time when the latest determination result information is received, the past 1 millisecond includes the times 132 “2601”, “2243”, and “1799”. It is. Therefore, the ratio (Y ÷ X) is obtained as Y ÷ X = 0.5 from X = 10 + 5 + 5 = 20 and Y = 5 + 5.

  Next, the non-duplicate data increase prediction unit 130 multiplies the ratio (Y ÷ X) by the input data throughput (referred to as Z) to calculate non-duplicate data throughput (Y ÷ X) × Z. For example, assuming that Z = 20 megabytes, the non-overlapping data increase prediction unit 130 calculates (Y ÷ X) × Z = 0.5 × 20 megabytes = 10 megabytes.

  Next, the non-overlapping data increase prediction unit 130 solves the equation shown in FIG. FIG. 5 is an equation for predicting the future occurrence of non-redundant data.

  In FIG. 5, B_max bytes are the capacity of the buffer 170, and B_current bytes are the current holding amount of the buffer 170. Therefore, the solution is the time (T_max seconds) until the total size of the non-overlapping data blocks held in the buffer 170 reaches B_max bytes.

  For example, if B_max = 40 megabytes and B_current = 10 megabytes, the non-duplicate data increase prediction unit 130 calculates T_max = (40 megabytes-10 megabytes) / 20 megabytes / second = 1.5 seconds. Thus, the non-duplicate data increase prediction unit 130 calculates T_max seconds as information for predicting the future generation amount of non-duplicate data.

  The non-overlapping data increase prediction unit 130 may calculate T_max seconds using a size given in advance as the size of the input data block and a value given in advance as the input data throughput. In this case, the duplication determination unit 120 may not include the size of the input data block and the input data throughput in the determination result information.

  The mode control unit 140 stores the prediction information (T_max seconds) from the non-overlapping data increase prediction unit 130 based on the presence / absence of the prediction information (T_max seconds) and whether or not the current holding amount of the buffer 170 is zero. The operation mode of the apparatus 200 is determined. The operation modes are, for example, a power saving mode and a normal power mode. The power saving mode includes, for example, powering off the storage apparatus 200 itself (excluding the power control unit 210), shifting to the power saving mode for each disk unit 220, and powering off. Next, the mode control unit 140 notifies the storage apparatus 200 of the determined operation mode.

  In the storage device 200, the power control unit 210 switches the operation mode of the storage device 200 based on the operation mode notified from the mode control unit 140 of the storage control device 100. Since the technology for reducing the amount of power consumption in switching to the power saving mode is well known, description thereof is omitted.

  The disk unit 220 is composed of a single disk or a plurality of disks, and stores an input data block (non-overlapping data block) output from the buffer output control unit 150 of the storage control device 100 as a storage data block.

  In this embodiment, the storage control device 100 manages the correspondence between the storage destination address 164 and the reference source address 165, but the storage device 200 manages the correspondence between the storage destination address and the reference source address. May be. In this case, the storage control device 100 may add the reference source address 165 to the non-overlapping data block and output it to the storage device 200.

  Next, the operation of the present invention will be described in detail with reference to the drawings.

  First, the host system 300 transmits a write command. The storage control device 100 receives a write command transmitted from the host system 300.

  When the write command is received, the data dividing unit 110 of the storage control device 100 divides the input data included in the write command to generate an input data block and outputs it.

  Next, the duplication determination unit 120 receives the input data block and refers to the hash value management table 161 to determine duplication of the received input data block.

  FIG. 6 is a flowchart showing the operation of the overlap determination unit 120.

  First, the duplication determination unit 120 receives an input data block (step S1101).

  Next, the duplication determination unit 120 refers to the hash value management table 161 to determine whether or not the input data block is a non-duplication data block including data different from the data included in the storage data block. (Step S1102).

  If it is not a non-overlapping data block (NO in step S1102), the duplication determination unit 120 discards the input data block (step S1103).

  Subsequently, the duplication determination unit 120 updates the reference source management table 162 (step S1104). Then, the process proceeds to step S1107.

  If it is a non-overlapping data block (YES in step S1102), the duplication determining unit 120 stores the input data block in the buffer 170 as a non-duplicating data block (step S1105).

  Subsequently, the duplication determination unit 120 updates the hash value management table 161 and the reference source management table 162 (step S1106).

  Next, the duplication determination unit 120 calculates the input data throughput (step S1107).

  Next, the duplication determination unit 120 outputs the size of the input data block, the duplication determination result (non-duplication / duplication), and the input data throughput to the non-duplication data increase prediction unit 130 as determination result information (step S1108). .

  Next, the operation of the non-redundant data increase prediction unit 130 will be described with reference to FIG. FIG. 7 is a flowchart showing the operation of the non-redundant data increase prediction unit 130.

  First, the non-redundant data increase prediction unit 130 receives the size of the input data block, the result of the overlap determination, and the input data throughput (step S1301).

  Next, the non-overlapping data increase prediction unit 130 records the received input data block size, the result of the determination of duplication, and the time when they are received as a duplication determination history 131 (step S1302).

  Next, the non-redundant data increase prediction unit 130 calculates a ratio (Y) of a value (Y) obtained by summing all the data sizes of the input data blocks in the past 1 millisecond and a sum (Y) of all the data sizes of the non-duplicate data. ÷ X) is calculated (step S1303).

  Next, the non-redundant data increase prediction unit 130 multiplies the ratio (Y ÷ X) by the input data throughput Z to calculate non-redundant data throughput (Y ÷ X) × Z (step S1304).

  Next, the non-redundant data increase prediction unit 130 acquires the current holding amount (B_current bytes) of the buffer 170 (step S1305).

  Next, the non-overlapping data increase prediction unit 130 calculates T_max seconds based on the formula shown in FIG. 5 (step S1306).

  Next, the non-overlapping data increase prediction unit 130 outputs the calculated T_max seconds to the mode control unit 140 (step S1307).

  In the above description, the non-redundant data increase prediction unit 130 operates so as to output prediction information every time it receives a result of the overlap determination. You may make it notify prediction information to.

  Next, the operation of the mode control unit 140 will be described with reference to FIG. FIG. 8 is a flowchart showing the operation of the mode control unit 140.

  First, the mode control unit 140 restarts a timer (not shown) (sets the timer value to “0” and starts timing) (step S1401).

  Next, the mode control unit 140 checks whether or not the prediction information (T_max seconds) is output from the non-overlapping data increase prediction unit 130 (step S1402).

  If the prediction information (T_max seconds) is output (YES in step S1402), the process proceeds to step S1410.

  If the prediction information (T_max seconds) is not output (NO in step S1402), the mode control unit 140 checks whether or not the timer time has reached a predetermined T_idle seconds (step S1404).

  If T_idle seconds have been reached (YES in step S1404), the process proceeds to step S1440.

  If T_idle seconds have not been reached (NO in step S1404), the process returns to step S1402.

  In step S1410, the mode control unit 140 subtracts the startup time T_start seconds of the storage apparatus 200 from T_max seconds (step S1410).

  When the subtraction result is 0 or less (YES in step S1410), the mode control unit 140 notifies the power supply control unit 210 of the storage apparatus 200 to set the operation mode to the normal power mode (step S1411). Then, the process returns to step S1401.

  If the subtraction result exceeds 0 (NO in step S1410), the process returns to step S1402.

  In step S1440, the mode control unit 140 checks whether or not the current holding amount of the buffer 170 is 0 (step S1440).

  If the current holding amount is not 0 (NO in step S1440), the process proceeds to step S1411.

  When the current holding amount is 0 (YES in step S1440), the mode control unit 140 notifies the power supply control unit 210 of the storage apparatus 200 to set the operation mode to the power saving mode (step S1442). Then, the process returns to step S1401.

  With the operation described above, the mode control unit 140 notifies the operation mode. Next, the storage apparatus 200 switches between the power saving mode and the normal power mode based on the notified operation mode.

  When the operation mode is switched from the power saving mode to the normal power mode, the storage device 200 completes startup after T_idle seconds.

  The buffer output control unit 150 monitors the storage apparatus 200 and detects a state where the activation of the storage apparatus 200 is completed.

  Note that the mode control unit 140 may monitor the storage apparatus 200 to detect a state where the activation of the storage apparatus 200 is completed, and notify the buffer output control unit 150 of this. That is, the mode control unit 140 may collectively control the interface related to the operation mode and state of the storage apparatus 200.

  When the activation of the storage device 200 is completed, the buffer output control unit 150 outputs the non-redundant data block held in the buffer 170 to the disk unit 220 of the storage device 200. Next, when the output of the non-overlapping data block is completed, the buffer output control unit 150 clears the buffer 170. At the same time, the buffer output control unit 150 updates the hash value management table 161 and the reference source management table 162.

  The effect of the present embodiment described above is that sufficient power saving can be achieved in the deduplication storage.

  The reason is that the non-redundant data increase prediction unit 130 calculates prediction information corresponding to the future occurrence state of the non-duplicate data block, and the mode control unit 140 determines whether the storage device 200 is based on the presence / absence of the prediction information and the prediction information. This is because the operation mode is controlled.

[Second Embodiment]
Next, a second embodiment of the present invention will be described in detail with reference to the drawings. Hereinafter, the description overlapping with the above description is omitted as long as the description of the present embodiment is not obscured.

  FIG. 9 is a block diagram showing the configuration of the storage control apparatus 101 according to the second embodiment of the present invention.

  Referring to FIG. 9, the storage control apparatus 101 according to the present embodiment includes a duplication determination unit 120, a non-duplication data increase prediction unit 130, a mode control unit 140, a buffer output control unit 150, a management information storage unit 160, and a buffer 170. .

  In this embodiment, the input data block is given from an external device (not shown), and the duplication determination unit 120 receives and processes this input data block. For example, the input data block may be input data included in a write command transmitted by the host system 300 described in the first embodiment.

  The functions and operations of the other elements are the same as the corresponding elements in the first embodiment.

  The storage control apparatus 101 may be a storage control apparatus 102 configured by a general-purpose computer as shown in FIG.

  FIG. 10 is a block diagram illustrating a configuration of a storage control apparatus that causes a computer to execute predetermined processing according to a program in the present embodiment.

  Referring to FIG. 10, the storage device 102 includes a CPU (Central Processing Unit) 710, a disk device 720, a storage unit 730, and a communication unit 750.

  The duplication determination unit 120, the non-duplication data increase prediction unit 130, the mode control unit 140, and the buffer output control unit 150 of the storage control device 101 correspond to the CPU 710, the disk device 720, and the storage unit 730. In addition, the management information storage unit 160 and the buffer 170 of the storage control apparatus 101 correspond to the storage unit 730.

  The CPU 710 expands the program stored in the disk device 720, for example, in the storage unit 730, and based on the expanded program, the duplication determination unit 120, the non-duplication data increase prediction unit 130, the mode control unit 140, and the buffer output control unit The same processing as 150 is executed.

  The disk device 720 stores the program.

  The storage unit 730 stores the expanded program. Further, the storage unit 730 stores a hash value management table 161 and a reference source management table 162 in the same manner as the management information storage unit 160. Further, the storage unit 730 stores non-overlapping data blocks, similarly to the buffer 170.

  The communication unit 750 is included in the duplication determination unit 120, the mode control unit 140, and the buffer output control unit 150.

  Also in the first embodiment, similarly, the storage control device 100 may be a storage control device 102 configured by a general-purpose computer as shown in FIG.

  The effect of the present embodiment described above is that, in addition to the effect of the first embodiment, the data dividing unit 110 is unnecessary and the components of the storage control apparatus 101 can be reduced.

  This is because the duplication determination unit 120 processes data received from the outside.

  Each component described in each of the above embodiments does not necessarily have to be individually independent. For example, for each component, a plurality of components may be realized as one module, or one component may be realized as a plurality of modules. Each component is configured such that a component is a part of another component, or a part of a component overlaps a part of another component. Also good.

  Further, in each of the embodiments described above, a plurality of operations are described in order in the form of a flowchart, but the described order does not limit the order in which the plurality of operations are executed. For this reason, when each embodiment is implemented, the order of the plurality of operations can be changed within a range that does not hinder the contents.

  Furthermore, in each embodiment described above, a plurality of operations are not limited to being executed at different timings. For example, another operation may occur during the execution of a certain operation, or the execution timing of a certain operation and another operation may partially or entirely overlap.

  Furthermore, in each of the embodiments described above, a certain operation is described as a trigger for another operation, but the description does not limit all relationships between the certain operation and the other operations. For this reason, when each embodiment is implemented, the relationship between the plurality of operations can be changed within a range that does not hinder the contents. The specific description of each operation of each component does not limit each operation of each component. For this reason, each specific operation | movement of each component may be changed in the range which does not cause trouble with respect to a functional, performance, and other characteristic in implementing each embodiment.

  Each component in each embodiment described above may be realized by hardware, software, or a mixture of hardware and software, if necessary. May be.

  Further, the physical configuration of each component is not limited to the description of the above embodiment, and may exist independently, may exist in combination, or may be configured separately. May be.

A part or all of the above-described embodiment can be described as in the following supplementary notes, but is not limited thereto.
(Appendix 1)
Whether the input data block that is a candidate for data to be stored in the storage device is a non-overlapping data block including data different from the data included in any storage data block already stored in the storage device, A comparison between a hash value of data included in the input data block and a hash value of data included in each of the storage data blocks, and a determination unit for outputting the determination result of the determination;
A buffer holding the non-overlapping data block;
A management information storage unit that stores a hash value and a storage destination address in association with each storage data block and each non-overlapping data block;
Based on the determination result, the capacity of the buffer, and the current used capacity of the buffer, prediction information corresponding to prediction of a future occurrence state of the non-duplicated data is calculated, and the calculated prediction information is output. A duplicate data increase prediction unit;
When the storage device has been activated, the non-overlapping data block stored in the buffer is output to the storage device, and the output non-overlapping stored in the management information storage unit A buffer output control unit for updating the storage destination address corresponding to the data block;
A mode control unit that determines an operation mode of the storage device based on the presence / absence of output of the prediction information and the prediction information, and notifies the storage device of the determined operation mode. A storage control device that stores the non-duplicate data block in the buffer and adds a set of the hash value and the storage destination address corresponding to the stored non-duplicate data block to a management information storage unit.
(Appendix 2)
The non-redundant data increase prediction unit calculates a ratio between a value obtained by summing all the data sizes of the input data blocks that meet a predetermined condition and a value obtained by summing all the data sizes of the non-duplicate data, A non-redundant data throughput is calculated by multiplying the calculated ratio by the input data throughput. The storage control device according to appendix 1, wherein the storage control device outputs the prediction information.
(Appendix 3)
The input data block is of variable length;
The storage control apparatus according to appendix 1 or 2, wherein the duplication determination unit 120 outputs the size of the input data block corresponding to the determination result.
(Appendix 4)
The storage control device according to any one of appendices 1 to 3, wherein the non-redundant data increase prediction unit outputs the prediction information every time the determination result is received.
(Appendix 5)
The storage control apparatus according to any one of appendices 1 to 4, wherein the non-redundant data increase prediction unit notifies the prediction information at predetermined time intervals.
(Appendix 6)
Based on the operation mode, a power control unit that switches between a power saving operation mode and a normal power operation mode,
A data storage unit that stores the received non-overlapping data block as the stored data block;
A storage system comprising: the storage control device according to any one of appendices 1 to 5.
(Appendix 7)
A storage control device having a buffer and a management information storage unit,
Whether the input data block that is a candidate for data to be stored in the storage device is a non-overlapping data block including data different from the data included in any storage data block already stored in the storage device, Judging by comparing the hash value of the data included in the input data block and the hash value of the data included in each storage data block, and outputting the determination result of the determination,
Storing the non-duplicate data block in the buffer, adding the set of the hash value and the storage destination address corresponding to the stored non-duplicate data block to a management information storage unit;
Based on the determination result, the capacity of the buffer and the current used capacity of the buffer, calculate prediction information corresponding to the prediction of the future occurrence state of the non-duplicated data, and output the calculated prediction information,
When the storage device has been activated, the non-overlapping data block stored in the buffer is output to the storage device, and the output non-overlapping stored in the management information storage unit Update the destination address corresponding to the data block;
The storage control method which determines the operation mode of the said storage apparatus based on the presence or absence of the output of the said prediction information, and the said prediction information, and notifies the determined said operation mode to the said storage apparatus.
(Appendix 8)
In the calculation of the prediction information, the storage control device calculates a ratio between a value obtained by adding all the data sizes of the input data blocks that meet a predetermined condition and a value obtained by adding all the data sizes of the non-overlapping data. Calculate the non-overlapping data throughput by multiplying the calculated ratio by the input data throughput, and divide the buffer capacity by subtracting the current usage of the buffer by the non-overlapping data throughput. The storage control method according to appendix 7, wherein time is calculated.
(Appendix 9)
The storage control device according to appendix 7 or 8, wherein the input data block has a variable length and outputs a size of the input data block corresponding to the determination result.
(Appendix 10)
The storage control device according to any one of appendices 7 to 9, wherein the prediction information is output every time the determination result is received.
(Appendix 11)
The storage control device according to any one of appendices 7 to 10, wherein the prediction information is notified at predetermined time intervals.
(Appendix 12)
In a computer having a buffer and a management information storage unit,
Whether the input data block that is a candidate for data to be stored in the storage device is a non-overlapping data block including data different from the data included in any storage data block already stored in the storage device, Judging by comparing the hash value of the data included in the input data block and the hash value of the data included in each storage data block, and outputting the determination result of the determination,
Storing the non-duplicate data block in the buffer, adding the set of the hash value and the storage destination address corresponding to the stored non-duplicate data block to a management information storage unit;
Based on the determination result, the capacity of the buffer and the current used capacity of the buffer, calculate prediction information corresponding to the prediction of the future occurrence state of the non-duplicated data, and output the calculated prediction information,
When the storage device has been activated, the non-overlapping data block stored in the buffer is output to the storage device, and the output non-overlapping stored in the management information storage unit Update the destination address corresponding to the data block;
A program for executing a process of determining an operation mode of the storage device based on the presence / absence of output of the prediction information and the prediction information, and notifying the storage device of the determined operation mode.
(Appendix 13)
In calculating the prediction information, the ratio between the total value of the data sizes of the input data blocks that meet a predetermined condition and the total value of the data sizes of the non-overlapping data is calculated and calculated. A process of calculating a non-overlapping data throughput by multiplying the ratio by the input data throughput, and calculating a time by dividing a value obtained by subtracting the current usage of the buffer from the capacity of the buffer by the non-overlapping data throughput. The program according to appendix 12, wherein the program is executed by the computer.
(Appendix 14)
The program according to appendix 12 or 13, wherein the input data block has a variable length, and causes the computer to execute a process of outputting the size of the input data block corresponding to the determination result.
(Appendix 15)
The program according to any one of appendices 12 to 14, which causes the computer to execute a process of outputting the prediction information each time the determination result is received.
(Appendix 16)
The program according to any one of appendices 12 to 15, wherein the computer is caused to execute a process of notifying the prediction information every predetermined time.

DESCRIPTION OF SYMBOLS 100 Storage control apparatus 101 Storage control apparatus 102 Storage control apparatus 110 Data division part 120 Duplication determination part 130 Non-redundant data increase prediction part 131 Duplication determination history 132 Time 133 Size 134 Determination result 140 Mode control part 150 Buffer output control part 160 Management information Storage unit 161 Hash value management table 162 Reference source management table 163 Hash value 164 Storage destination address 165 Reference source address 170 Buffer 200 Storage device 210 Power supply control unit 220 Disk unit 300 Host system 663 Hash value 710 CPU
720 Disk device 730 Storage unit 750 Communication unit 900 Network

Claims (10)

Whether the input data block that is a candidate for data to be stored in the storage device is a non-overlapping data block including data different from the data included in any storage data block already stored in the storage device, A comparison between a hash value of data included in the input data block and a hash value of data included in each of the storage data blocks, and a determination unit for outputting the determination result of the determination;
A buffer holding the non-overlapping data block;
A management information storage unit that stores a hash value and a storage destination address in association with each storage data block and each non-overlapping data block;
Based on the determination result, the capacity of the buffer, and the current used capacity of the buffer, prediction information corresponding to prediction of a future occurrence state of the non-duplicated data is calculated, and the calculated prediction information is output. A duplicate data increase prediction unit;
When the storage device has been activated, the non-overlapping data block stored in the buffer is output to the storage device, and the output non-overlapping stored in the management information storage unit A buffer output control unit for updating the storage destination address corresponding to the data block;
A mode control unit that determines an operation mode of the storage device based on the presence / absence of output of the prediction information and the prediction information, and notifies the storage device of the determined operation mode. A storage control device that stores the non-duplicate data block in the buffer and adds a set of the hash value and the storage destination address corresponding to the stored non-duplicate data block to a management information storage unit. The non-redundant data increase prediction unit calculates a ratio between a value obtained by summing all the data sizes of the input data blocks that meet a predetermined condition and a value obtained by summing all the data sizes of the non-duplicate data, A non-redundant data throughput is calculated by multiplying the calculated ratio by the input data throughput. The storage control apparatus according to claim 1, wherein the storage control apparatus outputs the prediction information. The input data block is of variable length;
The storage control apparatus according to claim 1, wherein the duplication determination unit 120 outputs a size of the input data block corresponding to the determination result. The storage control apparatus according to claim 1, wherein the non-redundant data increase prediction unit outputs the prediction information every time the determination result is received. The storage control apparatus according to any one of claims 1 to 4, wherein the non-duplicate data increase prediction unit notifies the prediction information at predetermined time intervals. Based on the operation mode, a power control unit that switches between a power saving operation mode and a normal power operation mode,
A data storage unit that stores the received non-overlapping data block as the stored data block;
A storage system comprising the storage control device according to claim 1. A storage control device having a buffer and a management information storage unit,
Whether the input data block that is a candidate for data to be stored in the storage device is a non-overlapping data block including data different from the data included in any storage data block already stored in the storage device, Judging by comparing the hash value of the data included in the input data block and the hash value of the data included in each storage data block, and outputting the determination result of the determination,
Storing the non-duplicate data block in the buffer, adding the set of the hash value and the storage destination address corresponding to the stored non-duplicate data block to a management information storage unit;
Based on the determination result, the capacity of the buffer and the current used capacity of the buffer, calculate prediction information corresponding to the prediction of the future occurrence state of the non-duplicated data, and output the calculated prediction information,
When the storage device has been activated, the non-overlapping data block stored in the buffer is output to the storage device, and the output non-overlapping stored in the management information storage unit Update the destination address corresponding to the data block;
The storage control method which determines the operation mode of the said storage apparatus based on the presence or absence of the output of the said prediction information, and the said prediction information, and notifies the determined said operation mode to the said storage apparatus. In calculating the prediction information, the storage control device calculates a ratio between a value obtained by adding all the data sizes of the input data blocks that meet a predetermined condition and a value obtained by adding all the data sizes of the non-overlapping data. The non-overlapping data throughput is calculated by multiplying the calculated ratio by the input data throughput, and the value obtained by subtracting the current usage of the buffer from the capacity of the buffer is divided by the non-overlapping data throughput to obtain a time. The storage control method according to claim 7, wherein the storage control method is calculated. The storage control apparatus according to claim 7 or 8, wherein the input data block has a variable length and outputs a size of the input data block corresponding to the determination result. In a computer having a buffer and a management information storage unit,
Whether the input data block that is a candidate for data to be stored in the storage device is a non-overlapping data block including data different from the data included in any storage data block already stored in the storage device, Judging by comparing the hash value of the data included in the input data block and the hash value of the data included in each storage data block, and outputting the determination result of the determination,
Storing the non-duplicate data block in the buffer, adding the set of the hash value and the storage destination address corresponding to the stored non-duplicate data block to a management information storage unit;
Based on the determination result, the capacity of the buffer and the current used capacity of the buffer, calculate prediction information corresponding to the prediction of the future occurrence state of the non-duplicated data, and output the calculated prediction information,
When the storage device has been activated, the non-overlapping data block stored in the buffer is output to the storage device, and the output non-overlapping stored in the management information storage unit Update the destination address corresponding to the data block;
A program for executing a process of determining an operation mode of the storage device based on the presence / absence of output of the prediction information and the prediction information, and notifying the storage device of the determined operation mode.

Images