JP5287414B2 - Storage system, storage power consumption reduction method, program - Google Patents

Description

  The present invention relates to a storage system and a storage power consumption reduction method, and more specifically, a combination of a storage virtual allocation function (thin provisioning) and a HDD (Hard Disk Drive) power saving function (MAID; Massive Array of Idle Disks). In addition, the present invention relates to a technology for efficiently reducing power consumption according to the operation state by combining a semiconductor memory (for example, SSD; Solid State Drive) and an HDD and obtaining a higher power consumption reduction effect.

  In recent years, an increase in power consumption in data centers has become a problem. In particular, in continuous operation of a storage system that holds data, it is necessary to cope with an increase in power consumption due to an increase in the amount of data to be held, such as backup data, at an accelerated rate.

  As a typical power consumption reduction technique when using a storage system, there are a virtual allocation function (thin provisioning) and a power saving function (MAID). Among these, the virtual allocation function is a technology that allocates a virtual logical disk (LD) to a business server and allocates a practical amount from a virtual allocation pool that is built in advance according to the write. This is a function for obtaining a power saving effect by stopping the operation of all HDDs constituting a RAID group for a specific RAID (Redundant Array of Independent Disks) group (see Patent Document 1).

  According to the virtual allocation function, the storage capacity of the physical storage system allocated to the logical disk can be reduced at the start of operation, and the operation can be started with a small number of HDDs. After the start of operation, the storage administrator monitors the amount of data used by the server and adds one HDD at a time if necessary. According to this technology, it is only necessary to operate the minimum HDD required according to the operation status of the server, so that the power consumption of the storage system can be reduced.

JP 2000-293314 A JP 2003-108317 A JP-A-5-233155 JP 2008-242971 A

  However, according to the virtual allocation function, once the HDD is incorporated into the virtual allocation pool and the actual capacity is allocated to the HDD, it is prohibited to delete the HDD from the pool, and it is necessary to keep it operating at all times. . Therefore, even if no data is actually stored, it is not allowed to reduce the physical capacity of the HDD, and the HDD must continue to operate in the pool. is there.

  Further, according to the power saving function (MAID), when one block in a RAID group is accessed, it is necessary to operate all HDDs constituting the RAID group. For this reason, according to the power saving function (MAID), there is a problem that it is difficult to effectively obtain a power saving effect for a pool that holds data frequently used in business.

  The present invention has been made in view of the above problems, and a storage system and storage consumption that can effectively reduce storage power consumption by optimizing data arrangement according to data access frequency and the like. An object is to provide a power reduction method.

In order to solve the above problems, a storage system according to the present invention is a storage system that provides a virtual storage area to a host device, and includes first and second virtual allocation pools allocated to the virtual storage area, In response to the placement request, data rearrangement means for rearranging data between the first and second virtual allocation pools based on the access frequency, and as a result of the rearrangement, the second virtual allocation pool When the access frequency of the arranged data does not reach a predetermined threshold and the free capacity of the second virtual allocation pool is equal to or larger than the predetermined capacity, the HDD constituting the second virtual allocation pool is deleted a pool capacity reduction means for reducing the capacity of the second virtual allocation pool by, after the relocation, certain time, a If Seth does not occur, it has a HDD off means for waiting the operation of the HDD constituting the second virtual allocation pool is stopped a predetermined time, a configuration of a storage system comprising a.

A storage power consumption reduction method according to the present invention is a storage power consumption reduction method for a storage system that provides a virtual storage area to a host device, and the virtual storage area is based on an access frequency in response to a data relocation request. A data rearrangement stage for performing data rearrangement between the first and second virtual allocation pools allocated to the network, and the access frequency of the data allocated to the second virtual allocation pool as a result of the rearrangement is predetermined. If the threshold value is not satisfied and the free capacity of the second virtual allocation pool is equal to or larger than a predetermined capacity, the second virtual allocation pool is deleted by deleting the HDD constituting the second virtual allocation pool. a pool capacity reduction step to reduce the capacity, the after relocation, a certain time, if the access has not occurred, Serial has a HDD off step of waiting a predetermined time the operation of the HDD stops constituting a second virtual allocation pool, the configuration of the storage power consumption reduction method comprising.

  According to the present invention, it is possible to optimize the data arrangement according to the data access frequency and the like, and to reduce the excess storage area. Therefore, it is possible to effectively reduce storage power consumption.

1 is a block diagram showing a configuration example of a storage system according to an embodiment of the present invention. It is a figure which shows an example of the mapping table 12 with which the storage system by embodiment of this invention is provided. It is a figure which shows an example of the access count table 13 with which the storage system by embodiment of this invention is provided. It is a figure which shows an example of the access frequency management table 22 with which the storage system by embodiment of this invention is provided. It is a flowchart which shows the flow of operation | movement of the rearrangement request | requirement means 11 with which the host computer by embodiment of this invention is provided. It is a flowchart which shows the flow of operation | movement of the data rearrangement means 14 with which the storage system by embodiment of this invention is provided. It is a flowchart which shows the flow of operation | movement of the HDD data rearrangement means 15 with which the storage system by embodiment of this invention is provided. It is a flowchart which shows the flow of operation | movement of the I / O management means 16 with which the storage system by embodiment of this invention is provided. It is a flowchart which shows the flow of operation | movement of the time passage judgment means 17 with which the storage system by embodiment of this invention is provided. It is a flowchart which shows the flow of operation | movement of the pool capacity reduction means 18 with which the storage system by embodiment of this invention is provided. It is a flowchart which shows the flow of operation | movement of the pool capacity expansion means 19 with which the storage system by embodiment of this invention is provided. It is a flowchart which shows the flow of operation | movement of the standby HDD ON means 20 with which the storage system by embodiment of this invention is provided. It is a flowchart which shows the flow of operation | movement of the standby HDD off means 21 with which the storage system by embodiment of this invention is provided.

  FIG. 1 shows a configuration of an information processing system including a storage system 2 according to an embodiment of the present invention. This information processing system includes a host computer (host device) 1 and a storage system 2 that provides a virtual storage area to the host computer 1. Among these, the host computer 1 includes a file system 10 and a relocation request unit 11, and the arrangement request unit 11 is for requesting the storage system 2 to relocate data. Further, as will be described in detail below, the storage system 2 receives the request from the relocation request means 11 of the host computer 1 and transfers the data on the pool in block units from two viewpoints of access frequency and free capacity. It has a function to rearrange and reduce the pool capacity after rearrangement.

[Description of configuration]
Hereinafter, the configuration of the storage system 2 will be described in detail with reference to FIG.
The storage system 2 includes a virtual logical disk (LD) 3, a virtual allocation pool (memory) 4, a virtual allocation pool (HDD) 6, a standby virtual allocation pool (HDD) 8, a free HDD group 9, and a mapping table 12. , Access count table 13, data relocation means 14, HDD data relocation means 15, I / O management means 16, time passage judgment means 17, pool capacity reduction means 18, pool capacity expansion means 19, standby HDD on means 20, A standby HDD off means 21 and an access count management table 22 are provided.

  Here, the virtual logical disk 3 is constructed on the virtual allocation pool (memory) 4 and the virtual allocation pools (HDD) 6 and 8 as having a virtual capacity corresponding to the virtual storage area provided to the host computer 1. And assigned to the host computer 1. The virtual allocation pool (memory) 4 is constructed on a plurality of semiconductor memories 5. The virtual allocation pool (HDD) 6 and the virtual allocation pool (HDD) 8 are each constructed on a plurality of HDDs (Hard Disk Drives) 7. These virtual allocation pool (memory) 4 and virtual allocation pools (HDD) 6 and 8 provide virtual capacity allocated to the virtual storage area of the virtual logical disk 3 by the virtual allocation function (thin provisioning).

  As illustrated in FIG. 2, the mapping table 12 defines a correspondence relationship between an LD number, an LD block number, a pool number, and a pool block number, and data of a certain LD block number of a certain LD. Is a table holding information indicating which pool block of which pool is stored. As illustrated in FIG. 3, the access count table 13 defines a correspondence relationship between a pool number, a pool block number, and a count number (a count value indicating the number of accesses). It is a table holding information indicating how many times the data of the pool block has been accessed. As illustrated in FIG. 4, the access count management table 22 defines a correspondence relationship between a pool number, a pool block number, a previous count count, and an access count, and based on the access count table 13 or This table holds information indicating how many times access has been made to the data of a certain pool block in the pool after the previous rearrangement process.

  The relocation request unit 11 executes a process for requesting the data relocation unit 14 in the storage system 2 to relocate data, as shown in FIG. As shown in FIG. 6 as an example of the processing, the data rearrangement unit 14 refers to the access count table 13 and rearranges data in the virtual allocation pool (memory) 4 in units of blocks according to the access frequency or the like. Processing and control are executed. The data rearrangement unit 14 calls the HDD data rearrangement unit 15 to rearrange data to be rearranged in the HDD. Details of the processing by the data rearrangement unit 14 (the flow in FIG. 6) will be described later.

  As shown in FIG. 7 as an example of the processing, the HDD data rearrangement unit 15 is called by the data rearrangement unit 14 described above, and the data is allocated to the virtual allocation pool (HDD) 6 or the standby virtual allocation pool (HDD). 8 for executing the rearrangement to 8. Details of the processing by the data rearrangement means 15 (the flow in FIG. 7) will be described later. As shown in step S61 of FIG. 8, when an access is made to a certain pool block of a certain pool, the I / O management means 16 executes a process of increasing the count count of the access count table 13 of the block by 1. To do.

  As shown in an example of the processing in FIG. 9, the time lapse judging means 17 is a standby virtual allocation pool (HDD) when a predetermined fixed time has elapsed from the time recorded in the I / O management means 16. 8 is executed to cause the standby HDD off means 21 to stop the operation of the HDD constructing 8. As shown in FIG. 10 as an example of the processing, the pool capacity reduction means 18 selects one HDD in the pool, copies all data of the HDD to another HDD in the pool, By moving to the free HDD group 9, processing for reducing the pool capacity is executed.

  As shown in FIG. 11 as an example of the process, the pool capacity expanding means 19 executes a process for expanding the capacity of the pool by incorporating one HDD from the free HDD group 9 into the pool. As shown in FIG. 12 as an example of the processing, the standby HDD on means 20 selects the HDDs constituting the standby virtual allocation pool (HDD) 8 when the standby virtual allocation pool (HDD) 8 is not operating. The process for operating is executed. As shown in FIG. 13 as an example of the processing, the standby HDD off means 21 is a standby virtual allocation pool (HDD) when the HDD that constructs the standby virtual allocation pool (HDD) 8 is operating. 8 is executed to stop the operation of the HDDs constituting the system 8.

[Description of operation]
Next, the operation (storage power consumption reduction method) of the embodiment of the present invention will be described in detail.
In normal operation, the I / O management means 16 of the storage system 2 records the number of counts indicating how many times each block in which pool has been accessed in the access count table 13 as needed. That is, when a certain block on the virtual logical disk 3 is accessed during operation, the block count is incremented by 1 and recorded in the access count table 13 (step S61 in FIG. 8). At this time, if the actual data of the accessed block exists in the standby virtual allocation pool 8 (HDD) (step S62 in FIG. 8; Yes), the standby HDD on means 20 causes the standby virtual allocation pool 8 to Is operated (step S63 in FIG. 8), and the time is recorded (step S64 in FIG. 8).

  Here, when the data in the storage system 2 is relocated, the relocation request unit 11 of the host computer 1 issues a data relocation request to request the data relocation unit 14 of the storage system 2 to relocate the data. (Step S11 in FIG. 5). Upon receiving the data relocation request, the data relocation unit 14 requests the standby HDD on unit 20 to operate the HDDs constituting the standby virtual allocation pool (HDD) 8 (step S21 in FIG. 6). Then, the standby virtual allocation pool (HDD) 8 is made available. That is, the standby HDD-on means 20 responds to the request from the data relocation means 14 if the HDDs constituting the standby virtual allocation pool (HDD) 8 are not operating (step S101 in FIG. 12; No), the HDD is operated (step S102 in FIG. 12).

  Then, the data rearrangement unit 14 calculates access count management representing the access frequency for each block from the previous execution of data rearrangement to the present based on the count count recorded in the access count table 13, and the access count The data is stored in the management table 22 (step S22 in FIG. 6). In addition, the data rearrangement unit 14 updates the previous count number on the access count management table 22 to the same number as the count number on the access count table 13 (step S22 in FIG. 6).

  Subsequently, the data rearrangement unit 14 refers to the access count management table 22 and performs a series of processes related to data rearrangement. That is, the data rearrangement unit 14 moves the data of the block having a high access frequency to the virtual allocation pool (memory) 4 of the semiconductor memory. Further, based on the request from the block relocation unit 14, the HDD data relocation unit 15 moves the data with low access frequency to the virtual allocation pool (HDD) 6 constructed from the HDD, and the data that has not been accessed at all. Are placed in the standby virtual allocation pool (HDD) 8 constructed for standby from the HDD.

Here, the data rearrangement process will be described in more detail with reference to FIG.
The block rearrangement unit 14 sorts the access count management table 22 in descending order by the access count management (step S23 in FIG. 6), and refers to one block of data in the descending order of access count management (step S24 in FIG. 6). The block rearrangement unit 14 includes an unprocessed block in the virtual allocation pool (memory) 4, that is, a block that has not been rearranged (step S <b> 25 in FIG. 6; Yes), and the data of the referenced block Is not present in the virtual allocation pool (memory) 4 (step S26 in FIG. 6; No), the block data on the virtual allocation pool (memory) 4 that is the relocation destination of the referenced block data is transferred to the HDD. The HDD data rearrangement means 15 is requested to perform the rearrangement (step S27 in FIG. 6), thereby causing the rearrangement to the virtual allocation pool (HDD) 6 or the standby virtual allocation pool (HDD) 8. Then, the block rearrangement unit 14 rearranges the data of the block referred to in the above step S24 in the virtual allocation pool (memory) 4 (step S28 in FIG. 6).

  As described above, when there is an unprocessed block in the virtual allocation pool (memory) 4 and the data of the referenced block does not exist in the virtual allocation pool (memory) 4, the block relocation unit 14 refers to In place of the unprocessed blocks on the virtual allocation pool (memory) 4, the data on the processed blocks are rearranged on the virtual allocation pool (memory) 4 and the unprocessed blocks are relocated to the virtual allocation pool (HDD). Rearrange to 6 or 8. Then, the block relocation unit 14 reflects the result of this data relocation on the mapping table 12, the access count table 13, and the access count management table 22 (step S29 in FIG. 6).

  Subsequently, the data rearrangement unit 14 determines whether or not all the access count management table 22 has been referred to (step S31 in FIG. 6), and if not all has been referred to (step S31; No), the process proceeds to step S24. And the data on the access count management table 22 is again referred to one block in the descending order of access count management. Then, the data rearrangement unit 14 performs processing of all blocks on the virtual allocation pool (memory) 4 when there is no unprocessed block on the virtual allocation pool (memory) 4 (step S25; No). If completed, the HDD data rearrangement means 15 is requested to rearrange the data of the referenced block (step S30 in FIG. 6). Thereafter, the above steps S24 to S31 are repeated until all the access count management table 22 is referred to. As described above, data rearrangement is performed.

Next, processing related to pool capacity reduction (steps S32 to S36 in FIG. 6) will be described.
When all the blocks on the access count management table 22 are referred to (step S31 in FIG. 6; Yes), that is, when the data rearrangement is completed, the data rearrangement means 14 has free space in the virtual allocation pool (HDD) 6. Is larger than the predetermined capacity (step S32 in FIG. 6; Yes), the pool capacity reduction means 18 is requested to reduce the pool capacity of the virtual allocation pool (HDD) 6 (step S33 in FIG. 6).

  Subsequently, the data rearrangement unit 14 determines that the standby virtual allocation pool (HDD) 8 has a free capacity equal to or larger than the predetermined capacity (step S34 in FIG. 6; Yes), the pool capacity reduction unit 18 uses the standby virtual allocation pool. A request is made to reduce the pool capacity of the assigned pool (HDD) 8 (step S35 in FIG. 6). In response to these requests, the pool capacity reduction means 18 selects one HDD in the pool to be reduced, and copies all data in the HDD to another HDD in the pool (step S81 in FIG. 10). . Then, the pool capacity is reduced by moving the HDD selected in step S81 to the empty HDD group 9 (step S82 in FIG. 10).

As described above, after the data rearrangement is completed, when the free capacity is larger than a certain value in the virtual allocation pool (HDD) 6 or the standby virtual allocation pool (HDD) 8, the HDD not storing data is assigned to the free HDD group 9. Reduce the pool capacity by migrating to.
Finally, the data relocation unit 14 requests the standby HDD off unit 21 to stop the operation of the HDDs constituting the standby virtual allocation pool (HDD) 8 (step S36 in FIG. 6). Thereby, power consumption is reduced.
The series of flows in FIG. 6 has been described above.

Next, data relocation processing on the HDD executed by the HDD data relocation unit 15 in response to the request in step S27 of FIG. 6 will be described with reference to FIG.
The HDD data rearrangement means 15 completes all the block rearrangement processes in the virtual allocation pool (memory) 4 when data in the virtual allocation pool (memory) 4 is rearranged to the HDD or when data is rearranged. 6, it is called by the data relocation unit 14 in step S27 of FIG. 6 and receives a request regarding the relocation of data on the virtual allocation pools (HDDs) 6 and 8 from the data relocation unit 14.

  Upon receiving this request, the HDD data relocation unit 15 manages the access count of the data on the currently referenced block at or above a predetermined threshold (step S41 in FIG. 7; No), and the block Is not present on the virtual allocation pool (HDD) 6 (step S42 in FIG. 7; No), the data of the block is rearranged in the virtual allocation pool (HDD) 6 (step S43 in FIG. 7). Conversely, when the access count management is less than the predetermined threshold (step S41 in FIG. 7; Yes), and the block does not exist on the standby virtual allocation pool (HDD) 8 (step S47 in FIG. 7). No), the data of the block is rearranged in the standby virtual allocation pool (HDD) 8 (step S48 in FIG. 7).

  In this data rearrangement process, the HDD data rearrangement means 15 determines that the free capacity of the virtual allocation pool (HDD) 6 or the standby virtual allocation pool (HDD) 8 is less than a predetermined capacity (step S45 in FIG. 7; Yes, or step S50; Yes), the pool capacity expansion unit 16 is requested to expand each pool (step S46 or step S51 in FIG. 7). That is, if each pool does not have a certain amount of free space, a request is made to incorporate HDDs in the free HDD group 9 into the virtual allocation pool. In response to this request, the pool capacity expansion means 16 incorporates one HDD from the free HDD group 9 into the pool and expands the capacity of the pool (step S91 in FIG. 11).

  When a block in the storage 2 is accessed during normal operation, the access counter on the access count table 13 is incremented by 1 for the block accessed by the I / O management means 16 (step S61 in FIG. 8). When the accessed block exists on the standby virtual allocation pool (HDD) 8 (step S62 in FIG. 8; Yes), the standby virtual allocation pool (HDD) 8 is constructed by the standby HDD on means 20 After operating the existing HDD (step S63 in FIG. 8), the time is recorded (step S64 in FIG. 8).

  When it is determined that the predetermined time has elapsed from the recorded time (step S71 in FIG. 9; Yes), the time elapse determining means 17 operates the HDDs that construct the standby virtual allocation pool (HDD) 8. Is requested to the standby HDD off means 21. In response to this request, the standby HDD off means 21 stops the operation of the HDD constructing the standby virtual allocation pool (HDD) 8 (step S72 in FIG. 9), and then waits for a certain time (FIG. 9). Step S73). In this way, when the standby HDD off means 21 receives a request to stop the operation of the HDD, if the HDD is constructing a standby virtual allocation pool (HDD) 8 (step S111 in FIG. 13; Yes), the operation of the HDDs constituting the standby virtual allocation pool (HDD) 8 is stopped (step S112 in FIG. 13).

According to the above-described embodiment, the power saving function can be applied in an optimized state from the two viewpoints of pool availability and access frequency, and the following two effects can be obtained simultaneously.
The first effect of the present embodiment is that a larger power consumption reduction effect can be obtained by minimizing the number of HDDs that are operated according to the operation status. The reason is that when the amount of data stored in the pool is sufficiently small, copying is performed inside the pool so that HDDs that do not store data can be created, and removing HDDs from the pool reduces the number of HDDs that make up the pool. This is because the operation of the HDD can be stopped.

  The second effect of the present embodiment is that a power saving effect can be obtained with high efficiency even when the LD in the storage is frequently accessed in business. The reason is that data with high access frequency is relocated to the memory, and data with low access frequency is relocated to the standby pool, and the HDDs that make up the standby pool are normally stopped, and in rare cases only when there is access This is because the operation time of HDDs constituting the pool can be shortened by operating.

  Here, for reference, the effect of the existing technology is compared. A technique for storing only frequently accessed data in a semiconductor memory exists as an existing technique (Japanese Patent Laid-Open No. 5-233155). On the other hand, in this embodiment, the data stored in the HDD is also classified according to the access frequency, and the power saving function is applied to achieve higher storage power consumption reduction.

  Further, there is an existing technology for stopping the operation of an HDD that stores data with low frequency by dividing data stored in the HDD according to access frequency (Japanese Patent Laid-Open No. 2003-108317, Japanese Patent Laid-Open No. 2008-242971). With this existing technology, the number of HDDs in each pool with high / low frequency cannot be changed. In contrast, in this embodiment, the number of HDDs operating according to the operation status is minimized, and the pool is reconfigured to stop the operation of as many HDDs as possible, thereby obtaining a greater power consumption reduction effect. Making things possible.

  In the above embodiment, the present invention is expressed as a storage system and its operation (storage power consumption reduction method). However, a program for causing a computer to function as each means constituting the storage system according to the above embodiment, Alternatively, it can be expressed as a program for causing a computer to execute the operation (storage power consumption reduction method) of the storage system according to the above-described embodiment.

As mentioned above, although embodiment of this invention was described, this invention is not limited to the above-mentioned embodiment, A various deformation | transformation is possible in the range which does not deviate from the summary of this invention.
For example, in the above-described embodiment, referring to FIG. 1, the host computer 1 includes the relocation request unit 11, and the user of the host computer 1 starts the data relocation of the storage system 2. However, the present invention is not limited to this, and a relocation request means for periodically issuing a data relocation request is provided in the storage 2 so that the storage 2 itself executes data relocation voluntarily. Also good.

  In the above-described embodiment, the access frequency is monitored in units of blocks by referring to the access count table 13 in FIG. However, the configuration may be such that the relationship between the block and the file is maintained and the access frequency is monitored / relocated in units of files.

  The present invention can be widely applied to storage systems having virtual capacity.

  1; host computer (host device), 2; storage system, 3; virtual logical disk, 4; virtual allocation pool (memory), 5; semiconductor memory, 6; virtual allocation pool (HDD), 7; HDD, 8; Virtual allocation pool (HDD), 9; free HDD, 12; mapping table, 13; access count table, 14; data relocation means, 15; HDD data relocation means, 16; I / O management means, 17; Progress determination means 18; Pool capacity reduction means 19; Pool capacity expansion means 20; Standby HDD ON means 21; Standby HDD OFF means 22; Access count management table S11, S21-S36, S41-S51, S61 -S64, S71-S73, S81, S82, S91, S101, S102, S111, S112; Step.

Claims (13)

A storage system that provides a virtual storage area to a host device, the first and second virtual allocation pools allocated to the virtual storage area;
Data relocation means for performing data relocation between the first and second virtual allocation pools based on access frequency in response to a data relocation request;
As a result of the rearrangement, when the access frequency of the data arranged in the second virtual allocation pool is less than a predetermined threshold , and the free capacity of the second virtual allocation pool is greater than or equal to a predetermined capacity, Pool capacity reduction means for reducing the capacity of the second virtual allocation pool by deleting the HDD constituting the second virtual allocation pool;
HDD off means for stopping the operation of HDDs constituting the second virtual allocation pool and waiting for a certain period of time when no access occurs for a certain period of time after the rearrangement;
Storage system with Before Symbol data rearrangement means,
The data on the first and second virtual allocation pools are referred to in block units in the order of frequent access, and the data of the referenced block does not exist on the first virtual allocation pool, and the first When an unprocessed block exists in one virtual allocation pool, the data of the referenced block is relocated to the first virtual allocation pool instead of the unprocessed block, and the unprocessed block is Relocated to the virtual allocation pool
Result of performing the rearrangement of all the blocks on the first and second virtual allocation pool, if the free capacity of the second virtual allocation pool is equal to or greater than a predetermined volume, the second virtual allocation pool The storage system according to claim 1, wherein the pool capacity reduction means is requested to reduce the capacity. A pool capacity expansion means for expanding the capacity of the second virtual allocation pool;
The data rearrangement means includes
3. The pool capacity expansion unit according to claim 2, wherein when the free capacity of the second virtual allocation pool is less than a predetermined capacity during the rearrangement, the pool capacity expansion unit is requested to expand the capacity of the second virtual allocation pool. Storage system. 4. The storage system according to claim 2, wherein the first virtual allocation pool is constructed on a semiconductor memory, and the second virtual allocation pool is constructed on an HDD.   The storage system according to any one of claims 1 to 4, further comprising relocation request means for periodically issuing the data relocation request.   6. The storage system according to claim 2, wherein the storage system is configured to perform data rearrangement in units of files instead of data rearrangement in units of blocks. A storage power consumption reduction method for a storage system that provides a virtual storage area to a host device,
A data relocation stage for performing data relocation between the first and second virtual allocation pools allocated to the virtual storage area based on access frequency in response to a data relocation request;
As a result of the rearrangement, when the access frequency of the data arranged in the second virtual allocation pool is less than a predetermined threshold , and the free capacity of the second virtual allocation pool is greater than or equal to a predetermined capacity, A pool capacity reduction step of reducing the capacity of the second virtual allocation pool by deleting the HDD constituting the second virtual allocation pool;
An HDD off stage in which, if no access occurs for a certain period of time after the rearrangement, the operation of the HDDs constituting the second virtual allocation pool is stopped and waited for a certain period of time;
Storage power consumption reduction method including. In the previous Symbol data re-deployment phase,
The data on the first and second virtual allocation pools are referred to in block units in the order of frequent access, and the data of the referenced block does not exist on the first virtual allocation pool, and the first When an unprocessed block exists in one virtual allocation pool, the data of the referenced block is relocated to the first virtual allocation pool instead of the unprocessed block, and the unprocessed block is Relocated to the virtual allocation pool
Result of performing the rearrangement of all the blocks on the first and second virtual allocation pool, if the free capacity of the second virtual allocation pool is equal to or greater than a predetermined volume, the second virtual allocation pool The storage power consumption reduction method according to claim 7, wherein the capacity is reduced. And further comprising a pool capacity expansion step for expanding the capacity of the second virtual allocation pool,
In the data rearrangement step,
9. The storage power consumption reduction method according to claim 8, wherein in the rearrangement process, when the free capacity of the second virtual allocation pool is less than a predetermined capacity, the capacity of the second virtual allocation pool is expanded. The storage power consumption reduction method according to claim 8 or 9, wherein the first virtual allocation pool is constructed on a semiconductor memory, and the second virtual allocation pool is constructed on an HDD.   The storage power consumption reduction method according to claim 7, further comprising a relocation request step of periodically issuing the data relocation request.   12. The storage power consumption reduction method according to claim 8, wherein data rearrangement is performed in file units instead of data rearrangement in block units.   A program for causing a computer to execute each step of the storage power consumption reduction method defined in any one of claims 7 to 8.

Images