JP2015191443A - Storage system, control device, storage device, input/output control method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To contribute to suppressing the degradation of access performance.SOLUTION: A storage system according to the present invention includes a control device and a storage device. The storage device is provided with a hierarchical disk pool, a logical disk in which a physical area is virtualized, and input/output control means. The control device is provided with a control device management table in which information for identifying a data set, virtual addresses, logical addresses, and information for identifying the logical disk are correlated, and control means for accepting a request and operating in accordance with the request. The control means acquires a logical address based on a virtual address corresponding to a write request and the control device management table, reads in a data block corresponding to the logical address and a parity block corresponding to the data block, calculates a parity, and writes the data to be written and the parity to the logical address.

Description

  The present invention relates to a storage system, a control device, a storage device, an input / output control method, and a program.

  There is a case where a storage system including two or more storage apparatuses having different access performances and a control apparatus for controlling the storage apparatuses is configured. In such a storage system, it is preferable to store data efficiently and effectively use each storage device. In such a storage system, it is preferable to efficiently access data stored in the storage apparatus.

  Therefore, RAID (Redundant Arrays of Independent Disks) technology is widely used, in which logical disks allocated from two or more storage devices are divided into data blocks divided into one or two or more to be virtually one disk. ing. When a storage system is configured using RAID, it is common to guarantee the data by calculating the parity of each data block.

  Patent Document 1 describes a storage system in which a host recognizes two or more storage devices as one virtual storage device. In the storage system described in Patent Document 1, each storage device is configured by a volume group that is grouped according to a predetermined policy (reliability, cost, etc.). The storage system described in Patent Document 1 rearranges data when a user designates a volume in units of groups and designates a storage to which data is to be moved.

JP2013-033515A

  The disclosure of the above prior art document is incorporated herein by reference. The following analysis has been made from the viewpoint of the present invention.

  As described above, in the storage system described in Patent Document 1, it is preferable that data can be efficiently stored and accessed. Accordingly, there are cases where the data area of the storage apparatus is released as appropriate to improve the access performance of the storage apparatus. When the data area of the storage device is released, the data configuration of the data block including the data area is changed. Therefore, it is necessary to recalculate the parity of the data block.

  Here, consider a case where the storage apparatus accepts a release request for a partial data area of a data block, and recalculates the parity of the data block when the release request is accepted. In this case, input / output processing (I / O (Input / Output) processing) occurs for the calculation of the parity of the data block even for an area for which a release request has not been received. As a result, there is a possibility that the calculation load will increase unnecessarily in the storage apparatus. As a result, the access performance of the storage apparatus may be degraded.

  Further, if the recalculation of the parity is suspended until the data area having a predetermined size is released, an inconsistency between the actual data area and the data area guaranteed by using the parity occurs. As a result, an input / output process for the released data area occurs, which may deteriorate the access performance of the storage apparatus.

  Therefore, an object of the present invention is to provide a storage system, a control device, a storage device, an input / output control method, and a program that contribute to suppressing degradation of access performance.

According to a first aspect of the present invention, there is provided a storage system including a control device and one or more storage devices connected to the control device,
The storage device includes a hierarchical disk pool including one or more physical areas;
A logical disk virtualizing the physical area,
The control device includes a virtual disk to which one or more logical disks are allocated, and a control unit that accepts a request and operates in response to the request, and the control unit sets a first logical address When a release request is received for the corresponding first virtual address, the first data area corresponding to the first logical address is moved from the first physical area including the first data area to the first virtual area. Move to the second physical area, which has a lower capacity cost than the first physical area, the first data block includes the first logical address, the first data set includes the first data block, and When the second data block includes a second data block different from the first data block, and the second data block includes a second logical address, the control means corresponds to the second logical address. When a write request is received for the second virtual address, the second data block and the parity block in the first data set are read, the parity is recalculated, and the second logical address is A storage system is provided that writes data to be written and the recalculated parity.

According to a second aspect of the present invention, a virtual disk assigned with one or more logical disks,
Control means for accepting the request and operating in response to the request,
When the control unit receives a release request for the first virtual address corresponding to the first logical address, the control unit changes the first data area corresponding to the first logical address to the first data area. From the first physical area containing the first physical area to the second physical area having a capacity cost lower than that of the first physical area, the first data block containing the first logical address, and the first data When the set includes the first data block and a second data block different from the first data block, and the second data block includes a second logical address, the control means includes the second data block. When a write request is received for the second virtual address corresponding to the second logical address, the second data block and the parity block in the first data set are read. Inclusive, parity recalculates, said second logical address, and writes the write target data and the parity that are the recomputed, a control device is provided.

  According to the third aspect of the present invention, a hierarchical disk pool including one or more physical areas connected to a control apparatus, a logical disk virtualizing the physical areas, and a request from the control apparatus A logical address management table in which the input / output control means that operates according to the information, the information for identifying the logical disk, the information for identifying the physical area, the logical address, and the priority are associated with each other. When the control device receives a release request corresponding to the first logical address, the output control means acquires the priority corresponding to the first logical address based on the logical address management table, If the acquired priority is a second priority higher than the first priority, the priority is changed to the first priority and registered in the logical address management table. Storage device is provided.

According to a fourth aspect of the present invention, the method includes a control step that receives and operates a request, and in the control step, when a release request is received for the first virtual address corresponding to the first logical address, The first data area corresponding to the first logical address is changed from the first physical area including the first data area to the second physical area having a capacity cost lower than that of the first physical area. Moving, a first data block includes the first logical address, a first data set includes the first data block and a second data block different from the first data block; When the second data block includes the second logical address, when a write request is received for the second virtual address corresponding to the second logical address in the control step, The second data block and the parity block in the first data set are read, the parity is recalculated, and the write target data and the recalculated parity are written to the second logical address. An input / output control method is provided.
This method is linked to a specific machine called a control device that controls the storage destination of data.

According to a fifth aspect of the present invention, there is provided a program that is executed by a computer that controls a control device, accepts a request, executes an operating control process, and corresponds to the first logical address in the control process. When a release request is accepted for the first virtual address, the first data area corresponding to the first logical address is moved from the first physical area including the first data area to the first virtual area. To the second physical area having a capacity cost lower than that of the physical area, the first data block includes the first logical address, the first data set includes the first data block, and the first data block A second data block different from the first data block, and the second data block includes a second logical address, in the control process, When a write request is accepted for the second virtual address, the second data block and the parity block in the first data set are read, the parity is recalculated, and the second logical address A program for writing the write target data and the recalculated parity is provided.
The program can be recorded on a computer-readable storage medium. The storage medium may be non-transient such as a semiconductor memory, a hard disk, a magnetic recording medium, an optical recording medium, or the like. The present invention can also be embodied as a computer program product.

  According to each aspect of the present invention, there are provided a storage system, a control device, a storage device, an input / output control method, and a program that contribute to suppressing degradation of access performance.

It is a figure for demonstrating the outline | summary of one Embodiment. 1 is a block diagram showing an overall configuration of a storage system 1 according to a first embodiment. 3 is a diagram illustrating an example of a RAID management table 30. FIG. It is a figure which shows an example of the free area | region management table 31. FIG. 3 is a diagram illustrating an example of a logical address management table 10. FIG. It is a figure which shows an example of the data storage policy table. 5 is a flowchart showing an example of the operation of a RAID control means 32. 3 is a flowchart showing an example of the operation of the input / output control means 12. 6 is a flowchart showing an example of the operation of the data storage destination confirmation unit 13;

  First, an outline of an embodiment will be described with reference to FIG. Note that the reference numerals of the drawings attached to the outline are attached to the respective elements for convenience as an example for facilitating understanding, and the description of the outline is not intended to be any limitation.

  As described above, a storage system that contributes to suppressing degradation of access performance and contributes to reducing capacity cost is desired.

  Therefore, the storage system 1000 shown in FIG. 1 is provided as an example. The storage system 1000 includes a control device 1001 and one or more storage devices 1002. The control device 1001 includes a virtual disk 1011 and a control unit (control unit) 1012. The storage device 1002 includes a hierarchical disk pool 1021 and a logical disk 1022.

  The hierarchical disk pool 1021 includes one or more physical areas. The logical disk 1022 is a storage area that is a virtualized physical area.

  The virtual disk 1011 is a storage area to which one or more logical disks 1022 are allocated. The control means 1012 receives a request and operates in response to the request.

  It is assumed that the control unit 1012 receives a release request for the first virtual address corresponding to the first logical address. In that case, the control unit 1012 transfers the first data area corresponding to the first logical address from the first physical area including the first data area to the second capacity cost lower than that of the first physical area. Move to the physical area.

  Here, it is assumed that the first data block includes the first logical address. Further, it is assumed that the first data set includes a first data block and a second data block different from the first data block. Further, it is assumed that the second data block includes the second logical address.

  It is assumed that the control unit 1012 receives a write request for the second virtual address corresponding to the second logical address. In that case, the control means 1012 reads the second data block and the parity block in the first data set. Then, the control unit 1012 recalculates the parity, and writes the write target data and the recalculated parity to the second logical address.

  Therefore, when receiving a release request, the storage system 1000 stores unnecessary data in a storage destination that has reduced the capacity cost of the storage destination while suppressing recalculation of parity. Also, in a data set that includes a data block for which a release request has been received, if a write request is received for a data block that is different from the data block for which the release request has been received, an area for storing unnecessary data is not read. To control. Therefore, the storage system 1000 contributes to suppressing deterioration of access performance.

[First Embodiment]
The first embodiment will be described in more detail with reference to the drawings. In the following description, the control device 1001 is referred to as a virtualization controller. In the following description, the control means (control unit) 1012 is referred to as a RAID control means (RAID control unit).

  FIG. 2 is a block diagram showing the overall configuration of the storage system 1 according to this embodiment.

  The storage system 1 shown in FIG. 2 includes a storage apparatus A101 to a storage apparatus D104, an external storage 200, a virtualization controller 300, and a host computer 400. FIG. 2 is not intended to limit the configuration of the storage system 1 according to the present embodiment to the configuration shown in FIG. In the following description, the storage devices A101 to D104 are referred to as the storage device 100 when it is not necessary to distinguish between them.

  The storage apparatus 100 has a hierarchization function.

  The external storage 200 is connected to the storage apparatus 100 via a network. The external storage 200 provides an external volume that the storage device 100 uses as a storage area.

  The virtualization controller 300 is connected to the storage apparatus 100 via a SAN (Storage Area Network). In addition, the virtualization controller 300 manages the logical disk allocated from each storage device 100 as a virtual disk.

  The host computer 400 is connected to the virtualization controller 300 via the SAN. The host computer 400 is assigned a virtual disk from the virtualization controller 300 and can read and write to the virtual disk.

  Next, the internal configuration of the storage apparatus 100 will be described in detail.

  The storage apparatus 100 includes a logical address management table 10, a data storage policy table 11, an input / output control means (input / output control unit) 12, a data storage destination confirmation means (data storage destination confirmation unit) 13, and a hierarchical disk. A pool 120 and a logical disk 130 are included. Although not shown in FIG. 2, the storage apparatus B102, storage apparatus C103, and storage apparatus D104 have the same internal configuration as the storage apparatus A101, and include a logical disk 131, a logical disk 132, and a logical disk 133, respectively. In the following description, the logical disks 130 to 133 are described as the logical disk 130 when not particularly distinguished.

  The hierarchical disk pool 120 includes an SSD (Solid State Drive) 110, an SAS (Serial Attached SCSI (Small Computer System Interface)) 111, an SATA (Serial ATA (Advanced Technology Attachment)) 112, and an external volume 113. Consists of. In general, the access performance increases in the order of SSD 110> SAS 111> SATA 112> external volume 113. In general, the cost increases in the order of SSD 110> SAS 111> SATA 112. In general, the capacity increases in the order of SATA112> SAS111> SSD110.

  The logical disk 130 is a disk obtained by virtualizing a drive group in the hierarchical disk pool and the external volume 113.

  In the hierarchization function of the storage apparatus 100, the actual data is moved according to the access frequency and priority of the data on the logical disk 130 to optimize the capacity cost. Specifically, in the hierarchization function of the storage apparatus 100, the data substance of the data with high access frequency and high priority is moved to the SSD 110 and the SAS 111. Here, for example, the storage apparatus 100 may determine data with an access frequency exceeding a predetermined threshold as “data with a high access frequency”.

  On the other hand, in the hierarchization function of the storage apparatus 100, the data entity of the data with low access frequency and priority moves to the SATA 112 and the external volume 113. Here, for example, the storage apparatus 100 may determine data with an access frequency equal to or lower than a predetermined threshold as “data with low access frequency”.

  The external storage 200 is realized by a storage service provided by a cloud service or the like, or another storage device on-premises.

  The input / output control unit 12 operates in response to a request from the virtualization controller 300. When receiving an I / O (Input / Output) request from the virtualization controller 300, the input / output control unit 12 executes an input / output process on the physical area (SSD 110, SAS 111, SATA 112, external volume 113). Specifically, the input / output control unit 12 refers to the logical address management table 10 and determines which physical area the I / O request is for. Then, the input / output control means 12 executes input / output processing for the corresponding physical area.

  The input / output control unit 12 changes the priority of the logical address when receiving a priority change request. Specifically, the input / output control unit 12 refers to the logical address management table 10 and changes the priority of the corresponding logical address to “low”.

  The data storage destination confirmation unit 13 confirms whether there is data for changing the storage destination. For example, it is preferable that the data storage destination confirmation unit 13 confirms whether there is data for changing the storage destination at a predetermined time period. Specifically, the data storage destination confirmation unit 13 refers to the logical address management table 10 and the data storage policy table 11. Here, it is assumed that the priority of the logical address is low, but there is data not stored in the physical area set in the data storage policy table 11. In that case, the data storage destination confirmation unit 13 stores the data in the physical area set in the data storage policy table 11.

  Next, the internal configuration of the virtualization controller 300 will be described in detail.

  The virtualization controller 300 includes a RAID management table 30, a release area management table 31, a RAID control unit 32, and a virtual disk 301.

  The virtual disk 301 treats the logical disk 130 using a software RAID or the like as a single data set with blocks obtained by dividing data and a parity generated from the blocks, and divides them into a plurality of logical disks 130. Arranged.

  The RAID control unit 32 operates in response to a request from the host computer 400. When the RAID control means 32 accepts the release request, the RAID control means 32 records the virtual address from which the request came. Then, the RAID control means 32 makes a priority change request for the logical address corresponding to the virtual address for which the request came.

  Next, consider a case where the RAID control unit 32 receives a read request. First, when the requested virtual address is not released, the RAID control means 32 reads data from the logical address corresponding to the virtual address. On the other hand, when the requested virtual address is released, the RAID control unit 32 returns zero to the request source without reading data from the logical address corresponding to the virtual address.

  Next, when the RAID control means 32 accepts a write request, it reads the data block including the virtual address from which the request came and the data of the logical address of the parity block of the data block. Then, the RAID control means 32 recalculates the parity and writes the data and the new logical parity to the logical address.

  Note that there are a TRIM command, an UNMAP command, a WRITE SAME command, and the like as types of release requests for releasing a storage block that is no longer needed from the host computer 400. Further, the release request may be processing that is internally determined by the storage controller.

  Next, the RAID management table 30 will be described with reference to FIG.

  The RAID management table 30 includes a data set number, a virtual address, a logical address, and information for identifying a logical disk. Specifically, the data set number is information indicating a collection of data for parity calculation. The virtual address is a virtual address that is recognized by the host computer 400 and is a target of an I / O request. The information for identifying the logical disk is information for identifying the storage destination of the virtual address data. In FIG. 3, information for identifying logical disks is indicated by LD1 to LD4.

  In the storage system 1 shown in FIG. 2, a virtual disk is composed of four logical disks 130-133. Therefore, in the RAID management table 30 shown in FIG. 3, there are four blocks for one data set number. Parity blocks are allocated in order to the logical disks LD1 to LD4 so as not to concentrate on a specific logical disk. For example, when the logical address size is 4 KB and the block size is 128 KB, 32 logical addresses are allocated to each block. Here, in the storage system 1 according to the present embodiment, the virtual address and the logical address have a one-to-one correspondence. Therefore, 32 virtual addresses are assigned to each block.

  Next, the release area management table 31 will be described with reference to FIG.

  As shown in FIG. 4, the release area management table 31 includes a virtual address that has received a release request from the host computer 400.

  Next, the logical address management table 10 will be described with reference to FIG.

  The logical address management table 10 includes a logical disk number, a logical address, information indicating a storage destination for identifying a physical area, a physical address / mapping, and a priority. Here, the logical disk number means a number for identifying a logical disk. The logical address means an address that the virtualization controller 300 recognizes and uses to issue an I / O request. The storage destination means a storage destination for identifying a physical area in which the substance of data is stored. The physical address / mapping means information for identifying the location of data at the storage destination. The priority means a priority for determining a data storage destination.

  In the logical address management table 10 according to the present embodiment, the priority is initially set to “high” for each logical address. When the storage apparatus 100 receives the priority change request, the storage apparatus 100 changes the priority to “low”.

  Next, the data storage policy table 11 will be described with reference to FIG.

  The data storage policy table 11 is set including the priority set for the logical address and the storage destination of the data corresponding to the priority. In the data storage policy table according to the present embodiment, the priority is either “high” or “low”. When the priority is “high”, the data is preferably stored in the physical drives (SSD 110, SAS 111, SATA 112) of the storage apparatus 100. On the other hand, when the priority is “low”, it is preferable to store data in the external volume 113 of the storage apparatus 100. The user may be allowed to set the correspondence between the priority and the data storage destination.

  Next, the operation of the storage system 1 according to this embodiment will be described.

  FIG. 7 is a flowchart showing an example of the operation of the RAID control means 32.

  In step S101, the RAID control unit 32 operates upon receiving a request from the host computer 400, and determines whether the request is a release request or an I / O request.

  When the RAID control unit 32 receives the release request, the RAID control unit 32 registers the requested virtual address in the release area management table 31 (step S102). Then, the RAID control unit 32 refers to the RAID management table 30 and confirms the logical disk corresponding to the requested virtual address and the logical address (step S103). Then, the RAID control unit 32 requests the I / O control unit 12 of the storage apparatus to which the logical disk belongs to change the low priority of the logical address (step S104).

  On the other hand, when the RAID control unit 32 receives an I / O request, the RAID control unit 32 determines whether the received request is a read request or a write request (step S105). If the received request is a read request, the process proceeds to step S106. On the other hand, if the received request is a write request, the process proceeds to step S110.

  When the read request is received, the RAID control means 32 determines whether or not the requested virtual address is registered in the release area management table 31 (step S106).

  When the requested virtual address is registered in the release area management table 31 (Yes branch in step S106), the RAID control unit 32 does not read the logical address from the logical disk and sends it to the host computer 400. Zero is returned (step S107).

  On the other hand, when the requested virtual address is not registered in the release area management table 31 (No branch in step S106), the RAID control means 32 refers to the RAID management table 30 and the corresponding logical disk, The logical address is confirmed (step S108). Then, the RAID control unit 32 requests the input / output control unit 12 of the storage apparatus to which the corresponding logical disk belongs to read from the corresponding logical address (step S109).

  On the other hand, when a write request is received, the RAID control means 32 refers to the RAID management table 30 and corresponds to the data block to which the requested virtual address belongs and the parity block in the data set to which the data block belongs. The logical disk to be checked and the range of logical addresses are confirmed (step S110).

  Next, the RAID control unit 32 requests the input / output control unit 12 of the storage apparatus to which the corresponding logical disk belongs to read from the corresponding logical address (step S111).

  Next, the RAID control means 32 recalculates the parity based on the data block after writing, the data block before writing, and the parity block before writing (step S112).

  Next, the RAID control means 32 requests the input / output control means 12 to write the data block after writing and the recalculated parity block to the logical address of the logical disk read in step S111 ( Step S113).

  Then, the RAID control unit 32 refers to the release area management table 31 and deletes the virtual address from the release area management table 31 when the virtual address requested to be written is registered (step S114).

  Next, the operation of the input / output control means 12 will be described.

  FIG. 8 is a flowchart showing an example of the operation of the input / output control means 12.

  In step S201, when the input / output control unit 12 receives a request from the RAID control unit 32, the input / output control unit 12 determines whether the received request is a low priority change request or an I / O request.

  When the input / output control means 12 receives the low priority change request, the input / output control means 12 refers to the logical address management table 10 and changes the priority of the requested logical address to “low” ( Step S202).

  On the other hand, when the input / output control unit 12 receives an I / O request, the input / output control unit 12 refers to the logical address management table 10 and stores the data storage destination corresponding to the requested logical address and the physical address. Address / mapping is confirmed (step S203). Then, the input / output control unit 12 performs I / O on the physical address / mapping of the confirmed storage destination (step S204). When the write process is executed, the input / output control unit 12 refers to the logical address management table 10, and when the priority of the logical address of the write destination is “low”, the priority is set to “high”. Change (step S205).

  Next, the operation of the data storage destination confirmation unit 13 will be described.

  FIG. 9 is a flowchart showing an example of the operation of the data storage destination confirmation unit 13.

  The data storage destination confirmation unit 13 refers to the logical address management table 10 and the data storage policy table 11, and confirms whether the storage destination of each logical address is the storage destination according to the policy (step S301). . Here, it is preferable that the data storage destination confirmation unit 13 operates at a predetermined time period.

  When there is a logical address that is not the storage destination according to the policy, the data storage destination confirmation unit 13 moves the data to the storage destination that is the policy, and stores the storage destination of the logical address management table 10 and the physical address. / Mapping is updated (step S302).

  As described above, when receiving a data release request, the storage system 1 according to the present embodiment stores unnecessary data in a storage destination with a low capacity cost while suppressing parity recalculation. In the storage system 1 according to this embodiment, it is assumed that a write request to another block is received for a data set to which a block including an area whose priority is set to “low” belongs. In this case, the storage system 1 according to the present embodiment performs control so that the area whose priority is set to “low” is not read from the logical disk. Therefore, the storage system 1 according to the present embodiment contributes to suppressing degradation of access performance and contributes to reducing capacity cost.

  In the above embodiment, the case where the priority of the logical address is set to “high” or “low” has been described as an example. However, this is not intended to limit the priority of logical addresses to two levels in the storage system 1 according to the present invention. In the storage system 1 according to the present invention, the logical address priority may be set to three or more levels.

  A part or all of the above-described embodiment can be described as in the following supplementary notes, but is not limited thereto.

  (Supplementary Note 1) Same as the storage system according to the first aspect.

  (Supplementary Note 2) The storage apparatus further includes a logical address management table in which information for identifying the logical disk, information for identifying the physical area, a logical address, and a priority are associated with each other, and the input / output control The means obtains a logical address corresponding to the virtual address for which the release request has been received, a second priority lower than the first priority associated with the logical address, and the logical address. The storage system according to appendix 1, wherein the logical address management table is updated in association.

  (Supplementary note 3) The storage system according to supplementary note 2, wherein the storage device further includes a data storage policy table in which the priority is associated with a storage destination of data corresponding to the priority.

  (Additional remark 4) The said control apparatus is further provided with the release area management table in which the said virtual address which received the release request was registered, and the said control means is set to the virtual address corresponding to the said release request, when a release request is received. The storage system according to appendix 3, wherein the logical address corresponding to the virtual address is requested to change the priority, and the virtual address corresponding to the release request is registered in the release area management table. .

  (Additional remark 5) When the said control means receives the read request, when the virtual address corresponding to the said read request is registered into the said release area management table, it reads data from the logical address corresponding to the said virtual address The storage system according to appendix 4, wherein when the virtual address is not registered in the release area management table, data is read from a logical address corresponding to the virtual address.

  (Additional remark 6) It is as the control apparatus which concerns on the said 2nd viewpoint.

  (Additional remark 7) It is as the storage apparatus concerning the said 3rd viewpoint.

  (Additional remark 8) It is as the input-output control method which concerns on the said 4th viewpoint.

  (Additional remark 9) It is as the program which concerns on the said 5th viewpoint.

  The disclosure of the above patent document is incorporated herein by reference. Within the scope of the entire disclosure (including claims) of the present invention, the embodiment can be changed and adjusted based on the basic technical concept. Further, various combinations or selections of various disclosed elements (including each element of each claim, each element of each embodiment, each element of each drawing, etc.) are possible within the framework of the entire disclosure of the present invention. is there. That is, the present invention of course includes various variations and modifications that could be made by those skilled in the art according to the entire disclosure including the claims and the technical idea. In particular, with respect to the numerical ranges described in this document, any numerical value or small range included in the range should be construed as being specifically described even if there is no specific description.

1,1000 storage system 10 logical address management table 11 data storage policy table 12 input / output control means (input / output control unit)
13 Data storage location confirmation means (data storage location confirmation unit)
30 RAID management table 31 Release area management table 32 RAID control means (RAID control unit)
100, 1002 Storage device 101 Storage device A
102 Storage device B
103 Storage device C
104 Storage device D
110 SSD
111 SAS
112 SATA
113 External volume 120, 1021 Hierarchical disk pool 130-133, 1022 Logical disk 200 External storage 300 Virtualization controller 301, 1011 Virtual disk 400 Host computer 1001 Controller 1012 Control means (control unit)

Claims (9)

A storage system including a control device and one or more storage devices connected to the control device,
The storage device
A hierarchical disk pool including one or more physical areas;
A logical disk that virtualizes the physical area;
With
The controller is
A virtual disk to which one or more logical disks are allocated;
Control means for accepting the request and operating in response to the request,
When the control unit receives a release request for the first virtual address corresponding to the first logical address, the control unit changes the first data area corresponding to the first logical address to the first data area. From the first physical area including the second physical area having a capacity cost lower than that of the first physical area,
The first data block includes the first logical address, the first data set includes the first data block and a second data block different from the first data block, and the second data If the block contains a second logical address:
When the control unit receives a write request for the second virtual address corresponding to the second logical address, the control unit reads the second data block and the parity block in the first data set. The storage system recalculates the parity and writes the write target data and the recalculated parity to the second logical address. The storage apparatus further includes a logical address management table in which information for identifying the logical disk, information for identifying the physical area, a logical address, and a priority are associated with each other.
The input / output control means acquires a logical address corresponding to the virtual address for which the release request has been received, a second priority lower than the first priority associated with the logical address, and the logical The storage system according to claim 1, wherein the logical address management table is updated in association with an address.   The storage system according to claim 2, wherein the storage apparatus further includes a data storage policy table in which the priority and a storage destination of data corresponding to the priority are associated with each other. The control device further includes a release area management table in which the virtual address that has received the release request is registered,
When the control unit receives the release request, the control unit requests the logical address corresponding to the virtual address to change the priority based on the virtual address corresponding to the release request, and responds to the release request. The storage system according to claim 3, wherein a virtual address to be registered is registered in the release area management table.   When the read request is received, the control means does not read data from the logical address corresponding to the virtual address when the virtual address corresponding to the read request is registered in the release area management table. 5. The storage system according to claim 4, wherein, when an address is not registered in the release area management table, data is read from a logical address corresponding to the virtual address. A virtual disk assigned with one or more logical disks;
Control means for accepting the request and operating in response to the request,
When the control unit receives a release request for the first virtual address corresponding to the first logical address, the control unit changes the first data area corresponding to the first logical address to the first data area. From the first physical area including the second physical area having a capacity cost lower than that of the first physical area,
The first data block includes the first logical address, the first data set includes the first data block and a second data block different from the first data block, and the second data If the block contains a second logical address:
When the control unit receives a write request for the second virtual address corresponding to the second logical address, the control unit reads the second data block and the parity block in the first data set. The controller that recalculates the parity and writes the write target data and the recalculated parity to the second logical address. A hierarchical disk pool including one or more physical areas;
A logical disk that virtualizes the physical area;
Input / output control means that operates in response to a request from the control device;
A logical address management table that associates information for identifying the logical disk, information for identifying the physical area, the logical address, and priority;
The input / output control unit obtains the priority corresponding to the first logical address based on the logical address management table when the control device receives a release request corresponding to the first logical address. If the acquired priority is a second priority higher than the first priority, the priority is changed to the first priority and registered in the logical address management table. apparatus. Including a control process that accepts and operates requests,
When a release request is received for the first virtual address corresponding to the first logical address in the control step, the first data area corresponding to the first logical address is changed to the first data area. From the first physical area including the second physical area having a capacity cost lower than that of the first physical area,
The first data block includes the first logical address, the first data set includes the first data block and a second data block different from the first data block, and the second data If the block contains a second logical address:
In the control step, when a write request is received for the second virtual address corresponding to the second logical address, the second data block and the parity block in the first data set are read. The input / output control method of recalculating the parity and writing the write target data and the recalculated parity to the second logical address. A program to be executed by a computer that controls a control device,
Accepts requests, executes operating control processes,
In the control process, when a release request is received for the first virtual address corresponding to the first logical address, the first data area corresponding to the first logical address is changed to the first data area. From the first physical area including the second physical area having a capacity cost lower than that of the first physical area,
The first data block includes the first logical address, the first data set includes the first data block and a second data block different from the first data block, and the second data If the block contains a second logical address:
In the control process, when a write request is received for the second virtual address corresponding to the second logical address, the second data block and the parity block in the first data set are read. , A program for recalculating the parity and writing the write target data and the recalculated parity in the second logical address.

Images