JP2007058777A - Disk array device and computer system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To flexibly construct constitution for a logic drive when expanding the capacity of a RAID. <P>SOLUTION: This disk array device has a capacity expansion detecting part 12 for detecting the addition of other HDD 4 to the first logic drive, a storage position determining part 13 for determining the HDD 4 to store a data of each block for reconstructing the RAID when the addition of the HDD 4 is detected, the first logic drive construction part 14 for reconstructing the first logic drive in a RAID level before the reconstruction, using a capacity corresponding to the HDD 4 having most blocks stored with the data as a capacity for the first logic drive after reconstructing each HDD 4, and the second logic drive construction part 15 for constructing the second logic drive in a residual area. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a disk array device, and more particularly to a disk array device having a function of expanding the capacity of a disk.

  There is RAID (Redundant Arrays of Independent (Independent) Disks) as a technique for improving the reliability and speed of a recording / reproducing apparatus using a physical drive such as an HDD (Hard Disk Drive).

This RAID is composed of functions such as striping, mirroring, parity check, ECC (Error Check and Correct), data and parity, and ECC distributed writing. RAID 0 to RAID 6, Furthermore, there are various levels such as RAID 0 + 1 and RAID 10/30/50.

  For example, RAID0 is also called striping, and stores data by dividing one piece of data. In particular, the processing speed can be increased by simultaneously reading and writing with a plurality of HDDs. In addition, RAID5 saves “parity”, which is information for restoration, while saving the data divided by striping to multiple HDDs, so even if one HDD breaks, it is saved in the remaining HDD. Since data can be restored by the parity being used, both speed and security can be achieved.

  By the way, in a state where a RAID is constructed with a predetermined number of HDDs in the past, when capacity expansion is performed by adding another HDD, the capacity of the added HDD is simply the increase in the capacity of the logical drive. Become.

  For example, in a state where RAID 0 is constructed with a single HDD with 80 GB capacity, when an additional HDD with the same 80 GB capacity is added to construct RAID 0, the additional capacity is also used as an existing logical drive. The capacity of one logical drive only increased from 80 Gbytes to 160 Gbytes.

  Accordingly, since the increased capacity is always incorporated into the existing logical drive, it is not possible to perform a flexible configuration of the logical drive such that a new logical drive is newly configured when the capacity is expanded.

In Patent Document 1, as an example of steps for flexibly configuring a logical drive, a disk array is configured using a plurality of virtual disk devices having the same capacity, including a plurality of possible disk devices generated on the same real disk device. As a result, a technique is disclosed in which a plurality of real disk devices having different capacities can be used to configure a large capacity disk array that is the same as or close to the total capacity of all the disk devices.
JP 2000-298556 A (page 8, FIG. 1)

  However, the technique disclosed in Patent Document 1 merely constructs a disk array using a plurality of virtual disks in order to efficiently use a plurality of real disks having different capacities. Therefore, even if the technique of the above-mentioned document 1 is used, when a disk array is already constructed in one logical drive, the configuration of the logical drive when capacity expansion is performed is flexibly constructed. I couldn't.

  Therefore, the present invention has been made to solve the above-described problem. When a disk array is already constructed in one logical drive, the capacity is expanded by maintaining the capacity of the disk array. It is an object of the present invention to provide a disk array device and a computer system that can use the increased amount as another logical drive and can flexibly construct the configuration of the logical drive.

  In order to solve this problem, the disk array device according to the present invention is different from the first logical drive in which a RAID is constructed at a predetermined RAID level by at least one physical drive having a plurality of block units. Detecting means for detecting whether or not another physical drive has been added, and when another physical drive is detected by the detecting means, it is stored in the plurality of blocks in order to reconstruct the RAID including the physical drive In the physical drive that has the largest number of blocks of data to be stored by the storage position determination means among the physical drives, the storage position determination means that determines in which physical drive the stored data is stored, Reconfigure each physical drive with the same capacity as the number of blocks stored in that physical drive. The capacity for the subsequent first logical drive is used, and the data stored in the plurality of blocks in the physical drive determined by the storage position determination means is sequentially stored for each block unit at the RAID level before reconstruction. And first logical drive construction means for reconstructing the first logical drive, and second logical drive construction means for constructing the second logical drive in an area where the first logical drive construction means is not provided in the physical drive. 2 logical drive construction means.

  In the computer system according to the present invention, is another physical drive added to the first logical drive in which a RAID is constructed at a predetermined RAID level by at least one physical drive having a plurality of block units? Detecting means for detecting whether or not, and when another physical drive is detected by the detecting means, which physical data is stored in the plurality of blocks in order to reconstruct the RAID including the physical drive. In the physical drive that has the largest number of blocks of data to be stored by the storage position determination means among the physical drives, the storage position determination means that determines whether to store in the drive is stored in the physical drive The first logical drive after the reconstruction of each physical drive with the same capacity as the number of blocks The data stored in the plurality of blocks in the physical drive determined by the storage position determination means is sequentially stored for each block unit at the RAID level before reconstruction, and the first logical First logical drive construction means for reconstructing a drive, and second logical drive construction means for constructing a second logical drive in an area where the first logical drive construction means is not provided in the physical drive It is characterized by having.

  It is possible to flexibly construct the logical drive configuration when expanding the RAID capacity.

(First embodiment)
Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a diagram showing a configuration of a disk array controller 100 according to an embodiment of the present invention. As shown in the figure, the disk array controller 100 includes a CPU (Central Processing Unit), firmware for disk array control executed by the CPU, ROM (Read Only Memory) in which various parameters are stored, and the like. Control processor 1, RAM (Random Access Memory) 2 used as a work area for disk array control, and HDDs 4-1, 4-2,. A plurality of ports 5-1, 5-2,... Connected to each of the HDDs 4-1, 4-2,. A disk controller 6 for controlling A host interface 7 for controlling communication with the host CPU of the computer system, and is constituted by a bus 8 as a transmission path connecting these. The HDDs 4-1, 4-2,... Are detachable from the respective ports 5-1, 5-2,.

  The control processor 1 functions as a RAID control unit 11, a capacity expansion detection unit 12, a storage location determination unit 13, a first logical drive construction unit 14, and a second logical drive construction unit 15 by firmware stored in the ROM. .

  The RAID control unit 11 divides and stores data into striping units as in RAID control, for example, RAID 0, and writes the same data in a plurality of HDDs using mirroring as in RAID 1, in addition to RAID 4, As in RAID 5, data parity is generated at the time of writing, and the data and parity are distributed and recorded in the same block of a plurality of HDDs 4-1, 4-2,.

  The capacity expansion detection unit 12 is a part that performs the confirmation of the total capacity of the physical disks of the HDDs 4-1, 4-2,. .

  The storage position determination unit 13 determines the position of data to be written by the RAID control unit 11 in consideration of the configured RAID level. That is, it is determined at which position of which physical disk to write. The RAID control unit 11 controls the disk controller 6 to write data at the position determined by the storage position determination unit 13.

  The first logical drive construction unit 14 detects the addition of a physical disk by the capacity expansion detection unit 12 and confirms the total capacity of the physical disk, and then the capacity of the first logical drive after reconstruction by the method described later. And the RAID control unit 11 causes the first logical drive to be reconstructed.

  The second logical drive constructing unit 15 constructs an area of the physical disk where the construction of the first logical drive is not performed as the second logical drive.

  The disk array controller 100 is connected to a bus in a computer system such as a personal computer, and can communicate with the host CPU of the computer system through the bus.

  Next, the operation of the disk array controller 100 of this embodiment will be described.

  FIG. 2 is a flowchart showing the operation of the disk array controller 100 during capacity expansion according to the embodiment of the present invention. FIG. 3 is a diagram showing an example of a RAID 0 disk array before capacity expansion in the embodiment of the present invention. FIG. 4 is a diagram showing an example of a RAID 0 disk array after capacity expansion after the embodiment of the present invention is implemented.

  In the RAID 0 disk array of FIG. 3, one logical drive (first logical drive) is constructed by two physical drives, HDD 4-1 and HDD 4-2. , Each of which is assigned to five blocks, and RAID 0 is constructed. Here, data of D0 to D9 is stored in each block as shown in FIG. That is, the HDD 4-1 stores data D0, D2, D4, D6, and D8, and the HDD 4-2 stores data D1, D3, D5, D7, and D9.

  Assuming FIG. 3 as an initial state, first, the disk controller of the present invention starts the subsequent processing by adding another physical drive in addition to the existing physical drive (step S10 in FIG. 2).

  That is, when the capacity expansion detection unit 12 determines that no other physical drive is added (No in step S10), the processing from step S20 onward is not performed until it is determined that this addition exists. Even if another physical drive is added, the processing after step S20 may not be performed until the user inputs capacity expansion.

  On the other hand, when the capacity expansion detection unit 12 determines that another physical drive has been added (Yes in step S10), the storage position determination unit 13 next determines the storage position to which data is written and stored ( Step S20). The determination of the storage position is performed in consideration of the RAID level that is constructed. In the case of FIG. 3, assuming that HDD 4-3, which is one more physical drive, is added to HDD 4-1 and HDD 4-2 (see FIG. 4), the order of data writing is HDD 4-1 and HDD 4-4. In consideration of the fact that 2 is constructed with RAID 0, write in order of HDD 4-1, HDD 4-2, HDD 4-3, HDD 4-1, HDD 4-2, HDD 4-3, HDD 4-1. decide.

  Therefore, the HDD 4-1 writes data D 0, D 3, D 6 and D 9 in the order of logical addresses, the HDD 4-2 writes data D 1, D 4 and D 7, and the HDD 4-3 stores D 2, D 5, It is determined that the data of D8 is written.

  Next, based on the storage location determined in step S20, the first logical drive constructing unit 14 determines the capacity of the first logical drive after reconstruction, which will be described later (step S30).

  This capacity is determined through the following two steps. (1) Among the physical drives, the storage position determination unit 13 identifies the physical drive that has the largest number of data blocks to be stored. (2) The same capacity as the capacity corresponding to the number of blocks stored in the physical drive is determined as the capacity for the first logical drive after reconstruction in each physical drive.

  That is, in FIG. 4, the first logical drive construction unit 14 first has the physical block having the largest number of data blocks stored by the storage location determination unit 13 among the physical drives HDD4-1 to HDD4-3. The drive is identified as the HDD 4-1. Next, the first logical drive construction unit 14 has the same capacity as four blocks (D0, D3, D6, D9) that are stored in the HDD 4-1 (200 GByte if one block is 50 GByte). The capacity (200 GB) is determined as the capacity of each physical disk for the first logical drive after reconstruction in HDD 4-1 to HDD 4-3.

  Next, the RAID control unit 11 writes (rebuilds) data, and constructs a new logical drive 1 (new first logical drive) with the three physical drives of HDD 4-1, HDD 4-2, and HDD 4-3. (Step S40).

  That is, the new first logical drive is constructed by sequentially storing the data of D0 to D9 in the predetermined physical drive determined by the storage position determination unit 13 for each block at the RAID level before the reconstruction. As a result, as shown in FIG. 4, each of the HDDs 4-1 to 4-3 has the same capacity as the four blocks (200 GB), that is, three times the capacity of the four blocks (600 GB) as a whole. As a logical drive, it is constructed with RAID0. As a result, the data of D0, D3, D6, and D9 are written in the order of logical addresses in the HDD 4-1, the data of D1, D4, and D7 are written in the HDD 4-2, and D2, D5 are written in the HDD 4-3. , D8 data is written. The area a of the HDD 4-2 and the area b of the HDD 4-3 play a complementary role for maintaining the logical drive capacity before the reconstruction, and the first logical drive is RAID 0 in the areas a and b. As long as this is maintained, no data will be written.

  Next, the RAID controller 11 constructs an area that has not been used as the first logical drive as a new logical drive (second logical drive) (step S50), and the processing of this flowchart ends. That is, the areas that have not been used as the first logical drive even after reconstruction in HDD 4-1 to HDD 4-3 (c to e in FIG. 4) are constructed as a logical drive different from the new first logical drive described above. This process ends.

  The area that is not used as the first logical drive refers to all of the so-called remaining areas in the HDD 4-1 to HDD 4-3 where the first logical drive construction unit 14 did not construct the logical drive. May be constructed as the second logical drive, or a part of the remaining area may be used as the second logical drive.

  By performing the above processing, a new logical drive (new first logical drive) that maintains the capacity of the conventional RAID 0 logical drive (first logical drive) using the newly added HDD. As a result, RAID 0 is constructed, and for the remaining capacity, another logical drive (second logical drive) can be constructed.

  That is, in FIG. 3, for example, in the case of RAID 0 configured by constructing one logical drive (first logical drive) having a capacity of 500 GB as a whole with HDD 4-1 and HDD 4-2, which are physical drives having a capacity of 250 GB, for example. Further, when an HDD 4-3 having a capacity of 250 GB is added, as shown in FIG. 4, D0 to D9 are allocated to three physical drives HDD4-1 to HDD4-3, and a new logical drive having a capacity of 600 GB is obtained. (New first new logical drive) is constructed, and another logical drive (second logical drive) is constructed with the remaining capacity (ce) of 150 GB.

  Therefore, according to the request of the user, it is possible to construct RAID 0 in the second logical drive as in the case of the new first logical drive, or it is possible to construct a RAID such as RAID 1, which is a different RAID level.

  Therefore, since the capacity of the HDD capacity expansion can be used as another logical drive while maintaining the capacity of the first logical drive before capacity expansion, the configuration of the logical drive can be flexibly constructed when expanding the capacity of the RAID. It becomes.

  Note that FIG. 3 shows RAID 0 as an example only, but the present invention is not limited to RAID 0, and can be applied to other RAID levels such as RAID 5 and further to RAID level changes as described below. Is possible.

(Modification of the embodiment of the present invention)
A modification of the embodiment of the present invention will be described below. In the first embodiment described above, the case of RAID 0 has been described. However, the present invention is not limited to RAID 0 as in the present modification, and can be implemented at other RAID levels. 5 and 6 show RAID 5 as one of modifications.

  FIG. 5 is a diagram showing an example of a RAID 5 disk array before capacity expansion in a modification of the embodiment of the present invention. In the RAID 5 disk array of FIG. 5, one physical drive (first logical drive) is configured with a capacity of 750 GB by three physical disks, HDD 4-1, HDD 4-2, and HDD 4-3. Assume that five blocks are allocated. Here, each block includes HDD 4-1, HDD 4-2, HDD 4-3, HDD 4-3, HDD 4-2, HDD 4-1, HDD 4-1. Assume that data P0-1, D2, P2-3, D3, P4-5,... D9 are stored.

  Here, P indicates a parity which is data restoration information. For example, P0-1 indicates the parity of D0 and D1. Note that the parity is derived by calculating an exclusive OR by the RAID control unit 11 based on the data to be restored.

  In the state of FIG. 5, if another physical drive unit (HDD 4-4) is added in addition to the existing physical drive (Yes in step S10 in FIG. 2), the storage location determination unit 13 sets the RAID level to The storage position is determined in consideration (step S20).

  In the case of FIG. 5, since it is constructed with RAID5, the order of writing is HDD4-1, HDD4-2, HDD4-3, HDD4-4, HDD4-4, HDD4-3, HDD4-2, Drives to be written are determined in the order of HDD 4-1, HDD 4-1.

  Accordingly, in FIG. 6, the first logical drive construction unit 14 first has the physical block having the largest number of blocks of data stored by the storage location determination unit 13 among the physical drives HDD4-1 to HDD4-4. The drive is identified as the HDD 4-4. Next, the first logical drive constructing unit 14 has a capacity (200 GByte because one block is 50 GByte) corresponding to four (P0-2, D3, D8, D9) which are the number of blocks stored in the HDD 4-4. The 200 GByte having the same capacity is determined as the capacity of each physical disk for the first logical drive after reconstruction in HDD 4-1 to HDD 4-4.

  Next, the RAID control unit 11 writes (rebuilds) the data, and a new logical drive 1 (new first logical drive) is created with the four physical drives of HDD 4-1, HDD 4-2, HDD 4-3, and HDD 4-4. ) Is constructed (step S40).

  That is, the new first logical drive is constructed by sequentially storing the data of D0 to D9 in the predetermined physical drive determined by the storage position determination unit 13 for each block at the RAID level before the reconstruction. Thereby, as shown in FIG. 6, each of HDD 4-1 to HDD 4-4 is constructed with RAID 5 as the first logical drive after reconstruction of 200 GB, that is, 800 GB as a whole. As a result, the data of D0, D5, and D6 is written in the order of logical addresses in the HDD 4-1, the data of D1, D4, and P6-8 are written in the HDD 4-2, and D2, P3 are written in the HDD 4-3. Data of −5 and D7 are written, and data of P0-2, D3, D8, and D9 are written to the HDD 4-4. The area c of the HDD 4-1, the area b of the HDD 4-2, and the area a of the HDD 4-3 play a complementary role for maintaining the logical drive capacity before the reconstruction. No data is written as long as the first logical drive maintains RAID5.

  Next, the RAID controller 11 constructs an area that has not been used as the first logical drive as a new logical drive (second logical drive) (step S50), and the processing of this flowchart ends. That is, an area (d to g in FIG. 6) that has not been used as the first logical drive after reconstruction in HDD4-1 to HDD4-4 is constructed as a logical drive different from the new first logical drive described above. This process ends.

  The area not used as the first logical drive refers to all of the so-called remaining areas in the HDD 4-1 to HDD 4-4 where the first logical drive construction unit 14 did not construct the logical drive. May be constructed as the second logical drive, or a part of the remaining area may be used as the second logical drive.

  By performing the above processing, a new logical drive (new first logical drive) that maintains the capacity of the conventional RAID 5 logical drive (first logical drive) by using the newly added HDD. As a result, RAID 5 is constructed, and regarding the remaining capacity, another logical drive (second logical drive) can be constructed.

  That is, in the case of RAID 5 configured by constructing one logical drive (first logical drive) with a capacity of 750 Gbytes in HDD 4-1 to HDD 4-3, which is a physical drive with a capacity of 250 Gbytes, for example, in FIG. When the HDD 4-4 having a capacity of 250 GB is added, as shown in FIG. 6, D0 to D9 are allocated to the four physical drives HDD4-1 to HDD4-4, and a new logical drive having a capacity of 800 GB (new The first new logical drive) is constructed, and another logical drive (second logical drive) is constructed with the remaining 200 GB capacity (d to g).

  Therefore, according to the user's request, it is possible to construct RAID 5 in the second logical drive in the same manner as the new first logical drive, or it is possible to construct a RAID such as RAID 1, which is a different RAID level.

  Therefore, since the capacity of the HDD capacity expansion can be used as another logical drive while maintaining the capacity of the first logical drive before capacity expansion, the configuration of the logical drive can be flexibly constructed when expanding the capacity of the RAID. It becomes.

1 is a diagram showing a configuration of a disk array controller 100 according to an embodiment of the present invention. 5 is a flowchart showing the operation of the disk array controller 100 during capacity expansion according to the embodiment of the present invention. The figure which shows the example of the disk array of RAID0 before the capacity expansion in embodiment of this invention. The figure which showed the example of the disk array of RAID0 after capacity expansion after giving embodiment of this invention. The figure which shows the example of the disk array of RAID5 before the capacity expansion in the modification of embodiment of this invention. The figure which showed the example of the disk array of RAID5 after capacity expansion after implementation of the modification of embodiment of this invention.

Explanation of symbols

1 Control processor 1
2 RAM2
3 Disk array 3
4-1, 4-2, 4-3, 4-4 HDD4-1
5-1, 5-2, 5-3 Port 5-1
6 Disk controller 6
7 Host interface 7
8 Bus 8
11 RAID controller 11
12 Capacity expansion detector 12
13 Storage position determination unit 13
14 First logical drive construction unit 14
15 Second logical drive construction unit 15
100 Disk array controller 100

Claims (6)

Detecting means for detecting whether or not another physical drive is added to the first logical drive in which a RAID is constructed at a predetermined RAID level by at least one physical drive having a plurality of block units;
Memory for determining which physical drive stores data stored in the plurality of blocks in order to reconstruct the RAID including the physical drive when another physical drive is detected by the detection means Positioning means;
Among the physical drives, the physical drive having the largest number of blocks of data to be stored by the storage position determining means has the same capacity as the capacity corresponding to the number of blocks stored in the physical drive. And the capacity for the first logical drive after reconstruction in the physical drive determined by the storage location determination means for each block unit at the RAID level before reconstruction. First logical drive constructing means for sequentially storing the first logical drive and reconstructing the first logical drive;
A disk array device comprising: a second logical drive constructing means for constructing a second logical drive in an area where the first logical drive constructing means is not provided in the physical drive.   The disk array device according to claim 1, wherein the detection unit is implemented when a RAID capacity expansion request is made.   In the second logical drive construction means, the area where the first logical drive construction means is not taken is all the remaining areas of the physical drive where the first logical drive construction means is not taken. The disk array device according to claim 1, wherein the disk array device is an area. Detecting means for detecting whether or not another physical drive is added to the first logical drive in which a RAID is constructed at a predetermined RAID level by at least one physical drive having a plurality of block units;
Memory for determining which physical drive stores data stored in the plurality of blocks in order to reconstruct the RAID including the physical drive when another physical drive is detected by the detection means Positioning means;
Among the physical drives, the physical drive having the largest number of blocks of data to be stored by the storage position determining means has the same capacity as the capacity corresponding to the number of blocks stored in the physical drive. And the capacity for the first logical drive after reconstruction in the physical drive determined by the storage location determination means for each block unit at the RAID level before reconstruction. First logical drive constructing means for sequentially storing the first logical drive and reconstructing the first logical drive;
A computer system comprising: a second logical drive constructing unit configured to construct a second logical drive in an area where the first logical drive constructing unit is not provided in the physical drive.   The computer system according to claim 1, wherein the detection unit is implemented when a RAID capacity expansion request is made.   In the second logical drive construction means, the area where the first logical drive construction means is not taken is all the remaining areas of the physical drive where the first logical drive construction means is not taken. The computer system according to claim 1, wherein the computer system is an area of

Images