JP3224741B2 - Data storage device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data storage device provided with a plurality of storage means such as a magnetic disk drive and functioning as a single storage device using these storage means.

[0002]

2. Description of the Related Art Conventionally, a disk array device has been known as a data storage device having a plurality of storage means.
The disk array device is a RAID (Redundant Arrays) devised by the University of California, Berkeley.
of Inexpensive Disks) architecture, and is provided with a plurality of magnetic disk devices (hereinafter referred to as DK). Among the “RAID architecture”, a disk array device adopting an architecture called RAID level 5 (hereinafter, referred to as a RAID 5 disk array device) has high reliability by having redundant data and a plurality of DKs in parallel. It realizes high throughput by accessing, and can exhibit high performance particularly in an application environment where random access to a relatively small amount of data frequently occurs, such as online transaction processing.

As shown in FIG. 4, the RAID 5 disk array device has a control unit 20 for controlling the entire device, a plurality of built-in DKs # 0 to 3, and a disk interface (hereinafter referred to as interface) for connecting these devices to each other. Are abbreviated as I / Fs) 20a to 20d, and a host I / F 20e connected to a host computer (not shown). The number of DKs provided in the RAID 5 disk array device is not particularly defined in the “RAID architecture”, but here, a case where four DKs # 0 to 3 are provided will be described as an example.

[0004] The RAID 5 disk array device configured as described above functions as if it were a single DK to the host computer. for that reason,
In the RAID5 disk array device, a disk address (hereinafter, referred to as a logical address) space viewed from the host computer as shown in FIG. 5A is allocated to a plurality of DK # 0 to DK # 3 as shown in FIG. In a logical address space.

The logical address space on DK # 0-3 is formed by a set of data blocks in which the number of DK # 0-3 is set as a row and the number of sectors (storage areas) of each DK is set as a column. Have been. The data block is data that is the minimum unit accessed by the host computer, and is, for example, a DK sector capacity of 51.
It has the same data amount as 2 bytes. In the set of data blocks in the logical address space, the data blocks 41a, 42a,
3a and these data blocks 41a, 42a, 43
parity block 4 which is an exclusive OR calculated from a
4a form a data string 40A. That is, one of the blocks constituting the data string 40A is the parity block 44a.

In this logical address space,
A data chunk 41 is formed by a set of a predetermined number of data blocks 41a, 41b, 41c, and 41d stored successively in one DK (same row). The data chunk 41 has a capacity 2 to 8 times (several kilobytes to several tens of kilobytes) the capacity accessed most frequently from the host computer. For example, four data blocks 41a, 41b, 41c, 41d , The capacity is 2 Kbytes.
In one data chunk 41, consecutive logical addresses (for example, d00 to d03) are sequentially given to the data blocks 41a, 41b, 41c, and 41d.

Further, in this logical address space, data chunks 45, 46, 4 arranged in the same column direction
A stripe 49 is formed by a set of 7, 48. However, the stripe 49 includes the parity chunk 46 composed of only the parity block. In a RAID5 disk array device, that is, an architecture called RAID level 5, parity chunks are stored in different rows for each stripe, that is, in different DKs.

An RA having such a logical address space
In the ID5 disk array device, a parity block is generated and stored for each of the data strings in one of the plurality of DK # 0 to DK # 3.
Even if K has failed, the data blocks on the failed DK can be reproduced from the remaining data blocks on the DK, so that the data reliability of the entire apparatus can be increased. Further, in the RAID5 disk array device, when reading a data block, a plurality of DKs # 0 to # 3 can be accessed in parallel, so that a high throughput can be obtained.

[0009]

SUMMARY OF THE INVENTION By the way, RAID5
In a disk array device, in response to a write request from a host computer, in addition to writing a data block to be written, it is necessary to update a parity block for the data block. Will drop. Especially when writing to data blocks that span different data chunks or different stripes,
The degree of decrease in throughput increases.

Here, a continuous data block 42 on a logical address spans two different data chunks.
The case where there is a write request for b, 43a (logical addresses d07, d08) will be described with reference to FIG. However, the data block 42b (logical address d
07) has a parity block 44b (logical address dp).
03), and the data block 43a (logical address d08) has a parity block 44a (logical address d).
p00). The data block 42b is DK # 1 and the data block 43a is DK # 1.
In # 2, the parity blocks 44a and 44b are DK # 3
, Respectively.

In response to a write request from the host computer, new write data (d0 in FIG.
7 ', d08') (Step 301; hereinafter, step is abbreviated as S), the controller 20 of the RAID 5 disk array device transfers the data to the cache / buffers 51, 52 of the controller 20. Store temporarily. The cache / buffer includes, for example, a RAM (Random Access Memory) provided in the control unit 20, and is used as a disk cache for storing a copy of data on DK # 0-3. It is also used as a temporary buffer for generating a parity block.

When the new write data is stored in the cache / buffers 51 and 52, the control unit 20 calculates the data block 42b of the logical address d07 from DK # 1 in order to calculate a new parity block for the write data. Read and cache / buffer 53
(S302), and reads the data block 43a of the logical address d08 from DK # 2 and stores it in the cache / buffer 54 (S303). Further, the control unit 20 reads the parity block 44b of the logical address dp03 and the parity block 44a of the logical address dp00 from DK # 3,
The data is stored in the buffers 55 and 56 (S304, S30
5).

The control unit 20 calculates the exclusive OR of all the data stored in the caches / buffers 51, 53 and 55, and stores the result in the cache / buffer.
The data is stored in the buffer 57 (S306). Similarly, the exclusive OR of all the data stored in the caches / buffers 52, 54, 56 is stored in the cache / buffer 58 (S307). The calculation results (dp00 ', dp03' in the figure) in the caches / buffers 57, 58 generated in this manner become new parity blocks.

For example, a data block before writing is d
Assuming that i and the parity block are dp and the other data blocks on which the parity block is calculated are dj and dk, respectively, the following relationship is established before writing.

Dp = di xor dj xor dk (1) (where “xor” is an exclusive OR operator)

If the data to be newly written is di 'and the new parity block is dp', the following relationship must be established.

Dp ′ = di ′ xor dj xor dk (2)

At this time, since dxor d = 0 and 0xor d = d, the following equation is derived from equation (2).

Dp '= (di' xor dj xor dk) xor 0 = di 'xor dj xor dj xor di xor di = (di xor dj xor DK) xor di xor di' = dp xor di xor di '... ... (3)

As a result, the cache / buffer 57,
The calculation result in 58 is proved to be each new parity block.

When a new parity block is stored in the cache / buffers 57 and 58, the control unit 20 stores the write data stored in the cache / buffer 51 in the D / D.
Write to logical address d07 on K # 1 (S30
8) Write the write data stored in the cache / buffer 52 to the logical address d08 on DK # 2 (S309). Also, the cache / buffers 57 and 58
Is written to the logical address dp03 and the logical address dp00 on DK # 3, respectively (S310, S311). That is, RAID5
In the disk array device, when a write request for a continuous data block on a logical address spans two data chunks from a host computer, the control unit 20 and each DK # respond to one write request.
0 to 3 times (S302 to S30)
5) and four writings (S308 to S311) must be performed.

This corresponds to a write request for a continuous data block extending over two stripes on a logical address (for example, a logical address d in FIG. 5B).
The same applies to the case where there is a request to write data to the memory cells 11 and d12 or d23 and d24). In this case, for one write request, between the control unit 20 and each of the DKs # 0 to # 3,
It is necessary to perform 2 to 4 readings and 2 to 4 writings.

That is, in the RAID 5 disk array device, when a write request is made to a continuous data block on a logical address over different data chunks or different stripes, a total of 4 to 8 is output to the built-in DK # 0-3. Times access must be made.
Therefore, in a system environment where such write requests frequently occur, a plurality of DK #
The high throughput obtained by parallel access to 0 to 3 is offset, and as a result, the throughput of the entire apparatus may be significantly reduced.

Accordingly, the present invention provides a data storage device capable of reducing the degree of decrease in throughput even when a write request to a continuous data block on a logical address occurs across different data chunks or different stripes. The purpose is to provide.

[0025]

SUMMARY OF THE INVENTION The present invention has been devised to achieve the above object, and a data storage device according to claim 1 has a plurality of storage areas for storing data. A plurality of storage means are provided, and different addresses are sequentially assigned to the respective storage areas of the plurality of storage means, and the respective storage areas of the plurality of storage means are represented by the number of these storage means as rows, and When the number of storage areas in is arranged as a column, a set of a predetermined number of storage areas arranged on the same row in the arrangement is defined as a data chunk, and a set of the data chunks arranged in the same column direction is defined as a stripe. Further, one storage area among the plurality of storage areas arranged on the same column is used as a parity for storing exclusive OR of data stored in other storage areas among the plurality of storage areas. If there is a storage area having a continuous address over different data chunks in one stripe, the storage areas are arranged on the same column. Is what you do.

According to the data storage device having the above configuration, in different data chunks in one stripe, storage areas assigned consecutive addresses are both arranged on the same column. The area is also the same. Therefore, when data requiring two storage areas is to be stored across different data chunks in one stripe, these data are stored in storage areas arranged on the same column. Therefore, writing of the exclusive OR according to the storage of the data may be performed for one parity area for the storage area on the same column. That is, it is not necessary to write an exclusive OR to a different parity area for each storage area.

Further, the data storage device according to claim 2 is
A plurality of storage means having a plurality of storage areas for storing data are provided, and different addresses are sequentially assigned to the respective storage areas of the plurality of storage means. When the number of storage means is set as a row, and the number of storage areas in these storage means is arranged as a column, a set of a predetermined number of storage areas arranged on the same row in the arrangement is defined as a data chunk, and The set of the data chunks arranged in the column direction is a stripe, and one storage area among a plurality of storage areas arranged on the same column is stored in another storage area among the plurality of storage areas. Is used as a parity area for storing exclusive OR of data, and if there are storage areas with consecutive addresses over different stripes, these storage areas And the respective parity areas for these storage areas are arranged on two different rows in the arrangement, and the storage area and the parity area are adjacent to each other on the two different rows. They are arranged.

According to the data storage device having the above-described configuration, the storage areas assigned consecutive addresses between different stripes and the respective parity areas for these storage areas are adjacent to each other on two different rows. Therefore, the storage area and the parity area adjacent to each other can be accessed at one time on each row. Therefore, when data requiring two storage areas is to be stored across two stripes, these data and the exclusive OR associated therewith need only be written once for each of two different rows. . That is, there is no need to write data or exclusive OR for each storage area or for each parity area.

[0029]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A data storage device according to the present invention will be described below with reference to the drawings. Here, a case will be described in which the present invention is applied to a RAID 5 disk array device.

The RAID 5 disk array device according to the present embodiment is configured as shown in FIG. However, unlike the conventional RAID 5 disk array device, a plurality of DK # 0-3
The upper logical address space is arranged as shown in FIG.

That is, in this logical address space, the order (arrangement direction) of the logical addresses given to each data block in these data chunks in data chunks adjacent to each other in one stripe is reversed. . For example, within stripe 1, data chunk 1
1 for each data block (d
00 to d03) and the data chunk 11
And the arrangement directions of the logical addresses (d04 to d07) for the respective data blocks in the adjacent data chunk 12 are opposite to each other.

Therefore, in this logical address space, between different data chunks in one stripe, each data block given a continuous logical address is arranged on the same data column. I have. For example, in one stripe 1, the data blocks 12a and 13a to which continuous logical addresses (d07 and d08) are provided over the adjacent data chunks 12 and 13 are both arranged on the same data string A. I have. At this time, the following relationship is established on the data string A.

Dp00 = d00 xor d07 xor d08 (4)

That is, for these data blocks 12a and 13a given successive logical addresses,
The same parity block 14a corresponds. Note that the arrangement of these data blocks is logical and does not indicate the physical arrangement of the sectors in the DK.

Further, in this logical address space, each data block given a continuous logical address over different stripes is stored in the adjacent D block.
K is stored. For example, between data chunks 13 and 15 located on different stripes 1 and 2, data blocks 13 b and 15 a to which continuous logical addresses (d 11 and d 12) are given are adjacent to DK # 2 and # 3, respectively. Is stored in

However, in the RAID 5 disk array device, parity chunks are stored in different DKs for each stripe. Therefore, in this logical address space, if there are data blocks 13b and 15a given continuous addresses (d11 and d12) over two stripes 1 and 2, these data blocks 13b and 15a and these data Block 13
b, 15a for each parity block 14
b and 16a are located on adjacent DKs # 2 and # 3, respectively. And these DK #
2, on # 3, the data block 13b and the parity block 16a, and the data block 15a and the parity block 14b are arranged so as to be adjacent to each other.

Next, in the RAID 5 disk array device having the above-described logical address space, an operation example in which a write request is made to a continuous data block on a logical address over two data chunks will be described with reference to FIG. This will be described with reference to FIG.

For example, there is a write request from the host computer to the data blocks 12a and 13a of the logical addresses d07 and d08, and new write data (d07 'and d08' in the figure) are transferred from the host computer together with this request. And (S101), R
The control unit 20 of the AID5 disk array device temporarily stores the data in cache / buffers 21 and 22 of the control unit 20.

When new write data is stored in the cache / buffers 21 and 22, the control unit 20 converts the data block 12a of the logical address d07 from DK # 1 to calculate a new parity block for these write data. Read and cache / buffer 23
(S102), and further reads the data block 13a of the logical address d08 from DK # 2 and stores it in the cache / buffer 24 (S103). Also, the control unit 20 reads the parity block 14a of the logical address dp00 from DK # 3 and
5 (S104).

Here, the control unit 20 calculates the exclusive OR of all the data stored in the caches / buffers 21, 22, 23, 24, 25, and stores the result in the cache / buffer 26. (S105). The calculation result (dp00 'in the figure) in the cache / buffer 26 thus generated becomes a new parity block.

After generating a new parity block, the control unit 20 transfers the write data stored in the cache / buffer 21 to the logical address d0 on DK # 1.
7, writing is performed instead of the data block 12a already stored in the cache / buffer 2 (S106).
2 is written to the data block 13 already stored at the logical address d08 on DK # 2.
Write in place of a (S107). The cache /
The new parity block stored in the buffer 26 is
Write in place of the parity block 14a already stored at the logical address dp00 on DK # 3 (S10
8). In this way, in the RAID 5 disk array device, writing to a continuous data block on a logical address is completed over two data chunks.

As described above, in the RAID 5 disk array device, even if there is a write request for continuous data blocks on a logical address over two data chunks, the parity blocks for these data blocks are the same. Therefore, in response to one write request from the host computer, the control unit 20 and each D
Three readings between K # 0 and K # 3 (S102-S1
04) and three write operations (S106 to S108), the write operation is completed. In other words, as before,
There is no need to perform four readings and four writings, and the writing operation is quicker than in the past. Therefore, even in a system environment in which such write requests frequently occur, it is possible to prevent the throughput of the entire apparatus from being significantly reduced, and as a result, the performance of the apparatus is improved.

Next, FIG. 3 shows an operation example in a case where a write request is made to a continuous data block on a logical address over two stripes in a RAID5 disk array device having the above logical address space. It will be described with reference to FIG.

Here, the data blocks 13b and 15a of the logical addresses d11 and d12 are sent from the host computer.
An example will be described in which a write request has been made to. However, the data block 13b (logical address d1)
1) includes a parity block 14b (logical address dp0).
3) corresponds to the data block 15a (logical address d).
12) has a parity block 16a (logical address dp).
04) correspond. In the control unit 20, the data block 13b and the parity block 16a,
Blocks stored in areas adjacent to each other, such as the data block 15a and the parity block 14b, can be read or written once each.

For example, there is a write request from the host computer to the data blocks 13b and 15a of the logical addresses d11 and d12, and new write data (d11 'and d12' in the figure) are transferred from the host computer along with this request. (S201), the control unit 20 temporarily stores the data in the cache / buffers 31 and 32 of the control unit 20.

When new write data is stored in the cache / buffers 31, 32, the control unit 20 calculates a new parity block for these write data from the DK # 2 and the data block 13b of the logical address d11. , And the parity block 16a of the logical address dp04 is read by one reading and stored in the cache / buffers 33 and 36 (S
202). Further, the control unit 20 reads the data block 15a of the logical address d12 from the DK # 3 and the parity block 14b of the logical address dp03 by one reading, and reads each of them from the cache / buffer 3.
4 and 35 (S203).

Here, the control unit 20 calculates the exclusive OR of all the data stored in the caches / buffers 31, 33, 35 and the cache / buffers 32, 3
The exclusive OR of all the data stored in the storages 4 and 36 is calculated, and the results are stored in the cache / buffers 37 and 38, respectively (S204, S20).
5). The cache / buffer 37 thus generated,
38 (dp03 ′, dp0 in the figure)
4 ′) becomes a new parity block.

When a new parity block is generated, the control unit 20 subsequently stores the contents stored in the cache / buffers 31 and 38 with the data block 13b already stored at the logical address d11 on DK # 2. Parity block 1 already stored at logical address dp04
6a, and is written by one write (S206). Further, the control unit 20 compares the contents stored in the cache / buffers 32 and 37 with the data block 15a already stored at the logical address d12 on the DK # 3 and the parity block already stored at the logical address dp03. 14b instead of 1
Writing is performed by writing twice (S207). In this way, in this RAID 5 disk array device, writing to a continuous data block on a logical address over two stripes is completed.

As described above, in this RAID 5 disk array device, if there are continuous data blocks on logical addresses over two stripes, these data blocks and the respective parity blocks for these data blocks Are always stored in two DKs, and in these DKs, data blocks and parity blocks are stored in adjacent areas. Therefore, in response to one write request from the host computer, two readings (S202, S203) and two writings (S206, S207) are performed between the control unit 20 and each of the DK # 0-3. Then, the write operation is completed. In other words, the write operation is quicker than the conventional two to four times of reading and two to four times of writing. Therefore, even in a system environment in which such write requests frequently occur, it is possible to prevent the throughput of the entire apparatus from being significantly reduced, and as a result, the performance of the apparatus is improved.

Note that, in the present embodiment, the present invention
Although the case where the present invention is applied to the D5 disk array device has been described, the present invention is not limited to this. For example, a RAID4 disk array device employing an architecture called RAID level 4 in “RAID architecture” may be used. Applicable. The RAID4 disk array device differs from the RAID5 disk array device in that the DK for storing the parity chunk is fixed. If the present invention is applied to such a RAID4 disk array device, the same effects as described above can be obtained.

In this embodiment, a disk array device having a magnetic disk device as a storage means has been described as an example, but the present invention is not limited to this. For example, the same effects as described above can be obtained by applying the present invention to a device having a magnetic tape device or a device having an optical disk device as a storage means. Further, in this embodiment, a case has been described in which a write request is made to two consecutive data blocks on a logical address over a data chunk or a stripe. For example, a write request to three to five consecutive data blocks is made. Even if there is, the same can be applied.

[0052]

As described above, in the data storage device of the present invention, even if there is a write request for a continuous data block on a logical address over different data chunks or different stripes, the data storage device is compared with the conventional one. Since the number of accesses to the means can be reduced, the degree of decrease in throughput can be reduced. In other words, even in a system environment where such write requests frequently occur, it is possible to prevent the throughput of the entire apparatus from being significantly reduced, and as a result, the effect of improving the performance of the apparatus is obtained. Can be

[Brief description of the drawings]

FIG. 1 is a layout diagram of data blocks in a logical address space in an example of a data storage device according to an embodiment of the present invention;

FIG. 2 is an explanatory diagram showing an operation example in a case where a write request is made to a continuous data block on a logical address over two data chunks in the data storage device having the logical address space shown in FIG. 1;

FIG. 3 is an explanatory diagram showing an operation example when a write request is made to a continuous data block on a logical address over two stripes in the data storage device having the logical address space shown in FIG. 1;

FIG. 4 is a block diagram illustrating a schematic configuration of an example of a conventional data storage device.

5A and 5B are layout diagrams of data blocks in a conventional data storage device, FIG. 5A is a layout diagram in a logical address space viewed from a host computer, and FIG. 5B is a logical address space in a plurality of magnetic disk devices. FIG.

FIG. 6 is an explanatory diagram showing an operation example when a write request is made to a continuous data block on a logical address over two data chunks in the conventional example.

[Explanation of symbols]

1, 2 stripes 11, 12, 13, 15 Data chunks 11a, 12a, 13a, 13b, 15a Data blocks 14a, 14b, 16a Parity blocks # 0, # 1, # 2, # 3 Magnetic disk drive

Claims (2)

(57) [Claims] A plurality of storage means having a plurality of storage areas for storing data; different addresses are sequentially assigned to respective storage areas of the plurality of storage means; When the number of storage areas is arranged as rows and the number of storage areas in the storage means is arranged as columns, a set of a predetermined number of storage areas arranged on the same row in the arrangement is data. A set of the data chunks arranged in the same column direction as a chunk is formed as a stripe, and one storage area among a plurality of storage areas arranged on the same column is replaced with another storage area in the plurality of storage areas. A data storage device as a parity area for storing an exclusive OR of data stored in a stripe, wherein consecutive addresses are assigned across different data chunks in one stripe A data storage device characterized in that if there is a storage area, the storage areas are arranged on the same column. 2. A plurality of storage means having a plurality of storage areas for storing data, and different addresses are sequentially assigned to respective storage areas of the plurality of storage means, When the number of storage areas is arranged as rows and the number of storage areas in the storage means is arranged as columns, a set of a predetermined number of storage areas arranged on the same row in the arrangement is data. A set of the data chunks arranged in the same column direction as a chunk is formed as a stripe, and one storage area among a plurality of storage areas arranged on the same column is replaced with another storage area in the plurality of storage areas. A data storage device serving as a parity area for storing exclusive OR of data stored in different stripes. If there is a storage area with a continuous address across different stripes, Storage areas, and respective parity areas for these storage areas are arranged on two different rows in the arrangement, and the storage area and the parity area are mutually different on the two different rows. A data storage device, which is arranged adjacent to each other.