JP2000076021A - Disk array control method and disk array device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and a device for controlling disk array with which data access is accelerated when the data is read by providing a data arranging method considering the characteristics of disk storage devices compos ing a disk array. SOLUTION: Devices having the disks of zone bit recording(ZBR) system are used for disk storage devices 12-1 to 12-5, a storage area for data to be originally stored is arranged on the outer peripheral side of disk and a storage area on a disk for redundant data is arranged on the inner peripheral side of disk than the storage area for such data. Concerning the ZBR system, data transfer speed is accelerated toward the outer periphery and by arranging the data to be originally stored on the outer peripheral side of disk storage devices 12-1 to 12-5, data transfer speed is accelerated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a disk array control method and a disk array device,
The present invention relates to a disk array control method and a disk array device suitable for use in an information distribution type server or a database server.

[0002]

2. Description of the Related Art With the recent development of computer systems, demands for external storage devices have been increasing. A disk array device has been developed to meet the demand. A disk array device achieves a large capacity by incorporating a plurality of disk storage devices, and speeds up recording or reproduction by moving each disk storage device in parallel, and adds redundant data to achieve reliability. I'm raising sex.

[0003] Such a disk array device has a RAID
(Redundant Arrays of Inexpensive Disks) by David A. Patterson and others at UC Berkeley's A Case for Redundant Arrays of Inexpensive Disks.
(RAID) ”UCB / CSD 87/391 (December 1987) introduces the basic concept.

[0004] In that paper, by adding redundant data (Parity) to data to be originally stored and storing the data, even if one of the disk storage devices fails, the redundant data can be used and read out. And a method for realizing high-speed data access by accessing disk storage devices in parallel.

However, regarding the method of arranging redundant data, as in the RAID-3 system shown in FIG.
Since one of the (N + 1) N + 1 units, for example, the disk storage device 12- (N + 1) is all used as the disk storage device for redundant data, the parallelism at the time of data reading is not N + 1 times but N times. By using a method of arranging redundant data randomly on the entire surface of each of the disk storage devices 12-1 to 12- (N + 1) as in the RAID-5 method shown in FIG. In the worst case when reading data, a full seek from the inner circumference to the outer circumference of the disk may occur, and the access speed may be reduced.

The RAID-3 system is striping in units of bits, while the RAID-5 system is striping in units of blocks, and each block shown in FIG. 9 indicates a data block.

When data stored in individual disk storage devices incorporated in a disk array device is divided into data to be stored and redundant data, data to be stored and redundant data are stored in individual disk storage devices. A conventional document does not mention a method of arranging storage devices in consideration of access characteristics.

[0008]

SUMMARY OF THE INVENTION It is an object of the present invention to provide a disk array which realizes a high-speed data access at the time of data reading by realizing a data arranging method in consideration of characteristics of a disk storage device constituting a disk array. An object of the present invention is to provide a control method and a disk array device.

[0009]

In order to solve the above-mentioned problem, a disk storage device having a zone bit recording type disk is used, and a data arrangement utilizing the characteristics is used. In other words, in the disk array control method according to the first aspect, the redundant data is added to the input data, and the input data and the redundant data are separately stored in a plurality of disk storage devices having a zone bit recording type disk. The control method is characterized in that a storage area of a disk for redundant data is arranged on an inner peripheral side of the disk with respect to a storage area of input data.

In the disk array control method according to the second aspect, in addition to the disk array control method according to the first aspect, the storage area of the input data is set in the usable area of the disk.
It is characterized by being arranged from the outer peripheral side.

In the disk array control method according to a third aspect, in addition to the disk array control method according to the first aspect, the method of arranging input data in the storage area of the input data is as follows. N + the total number of disk storage devices
1, when the numbers of the disk storage devices are 1 to N + 1, the disk storage devices of the number 1 are arranged in order from the disk storage device of the number 1 to the disk storage device of the number N + 1. The method of arranging redundant data in the storage area of the redundant data is performed by arranging the redundant data in a disk storage device in which the divided input data corresponding to the redundant data is not arranged. It is characterized by doing.

Further, in the disk array device according to the present invention, redundant data is added to input data, and zone data is added.
A disk array device for separately storing input data and redundant data in a plurality of disk storage devices having a recording type disk, wherein a storage area of the disk for redundant data is set at an inner circumference of the disk from a storage area of the input data. It is characterized in that it is arranged on the side.

According to a fifth aspect of the present invention, in addition to the disk array device of the fourth aspect, a storage area for input data is arranged from an outer peripheral side in a usable area of the disk. Features.

According to a sixth aspect of the present invention, in addition to the disk array device of the fourth aspect, the arrangement of the input data in the storage area of the input data is as follows: Assuming that the total number of devices is N + 1 and the numbers of the disk storage devices are 1 to N + 1, the disk storage devices are arranged in order from the disk storage device of number 1 to the disk storage device of number N + 1. And the redundant data is arranged in the storage area of the redundant data in the disk storage device in which the divided input data corresponding to the redundant data is not arranged. It is characterized by being arranged in a storage device.

The operation of the above means will be described. The input data to be originally stored is stored in the data area, and the redundant data is stored in a redundant data area different from the data area. In addition, the redundant data area is disposed on the inner peripheral side of the data area in a zone bit recording (hereinafter, referred to as "ZBR") type disk. The redundant data is essentially unnecessary data and is recovery data used when a disk storage device fails. That is, the data is not necessary at the time of normal reading.

When data arranged by the disk array control method of the present invention is read, usually only the reading of the data area needs to be performed, and therefore, the seek operation to the inner peripheral side with the redundant data is not performed. In the ZBR system, the data transfer speed is higher on the outer peripheral side, and the data area is arranged on the outer peripheral side of the disk storage device, so that the data transfer speed at the time of normal reading is higher. In addition, since the data to be stored is arranged in a lump, the data transfer speed is increased without redundant data, and the seek range is narrowed, so that the access time can be reduced. Increasing the data transfer speed means, conversely, reducing the data transfer time.

Since redundant data is allocated to a disk storage device in which data corresponding to the redundant data is not allocated, even if one of the disk storage devices fails, the original function of RAID is that data can be recovered. Do not lose.

[0018]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described with reference to FIGS. Here, FIG. 1 is a configuration diagram of an information distribution server, FIG. 2 is an example of data arrangement in a case where the information distribution server is configured by five disk storage devices, and FIG.
FIG. 4 is a flowchart of a process for securing a necessary area of the RAID controller, and FIG. 5 is a flowchart of a process at the time of writing by the RAID controller;
FIG. 6 is a flowchart of a process at the time of reading by the RAID controller, and FIG. 7 is a diagram showing a seek range at the time of reading. Note that components in the figure that have the same structure as the conventional technology are denoted by the same reference numerals.

First, a configuration example of the information distribution server will be described with reference to FIG. Inside the server, a host CPU 2, a memory 3, and a network adapter 4 for communicating with the outside of the server are connected by a host bus 1. In the case of a server equipped with a disk array, the RAID controller 5 is connected to the host bus 1 via a BIU (Bus Interface Unit) 6.

The BIU 6 has an interface function between the host bus 1 and the control CPU bus 7 and a bus bridge function. PC in this BIU6
A configuration including an I bus interface function is also conceivable. Instead of directly connecting via the BIU 6, another network adapter other than the network adapter 4 attached to the host bus 1 for information distribution is added, and the RAID controller 5 is connected via this network. A form of connection is also possible.

A BIU 6, a control CPU 8, a memory 9, a redundant data calculation circuit 10, and a disk controller 11 are connected to a control CPU bus 7 in the RAID controller 5. The memory 9 is used for a work area of the control CPU 8, various tables and lists, and is also used as a disk cache.
The redundant data calculation circuit 10 includes a disk storage device 12-1.
Used to calculate redundant data used for ~ 12- (N + 1). The redundant data is, for example, parity.

The RAID controller 5 has a host CPU
2, the data instructed to be written is divided into a certain size (hereinafter, referred to as "striping"), and redundant data is calculated using the striped data as one unit (stripe). By using the redundant data, even if one of the disk storage devices 12-1 to 12- (N + 1) fails and data is lost, the original lost data can be recovered. it can.

The RAID controller 5 and the disk storage devices 12-1 to 12- (N + 1) are connected by a disk bus 13 via a disk controller unit 11, and the redundant data calculated in the RAID controller 5 and the stripe data are stripped. The written data is written in parallel to each of the disk storage devices 12-1 to 12- (N + 1).

The disk storage device 12- of the present embodiment
Nos. 1 to 12- (N + 1) use a disk having a plurality of zones with different data transfer rates in the ZBR system. Under the condition that the recording density is constant and the rotation speed of the disk is constant, a larger number of sectors per track can be obtained on the outer periphery than on the inner periphery. ZB
The R system utilizes this property to divide the recording surface into a plurality of zones according to the distance from the center of the disk, and to make the outer zone use more sectors per track in the inner zone than the inner zone. This is a format method to increase the capacity.

In a disk storage device using a ZBR type disk, since the disk rotation speed is constant and the number of sectors is larger on the outer circumference than on the inner circumference, data is recorded on the outer circumference even if the data size is the same. Data is read faster than data recorded on the inner circumference. Taking advantage of such access characteristics, striped data is recorded on the outer peripheral side of the disk, and redundant data is recorded on the inner peripheral side of the disk from the data.

An example of data arrangement when used in five disk storage devices will be described with reference to FIG. SS1 to SS5 are 1
A set of striped data constituting a stripe, each of which is composed of four data. For example, SS1 is composed of data D1 to D4. The striped data is arranged in order from the disk storage device 12-1 to the disk storage device 12-5, and in each disk storage device, arranged in order from the outer periphery of the disk.

From P (D1 to D4) to P (D37 to D4)
0) is redundant data corresponding to the striped data. For example, P (D1 to D4) is SS1 redundant data. The redundant data is arranged collectively from the inner circumference side in order from the data arrangement. From P (D1 to D4) to P (D17 to D
20) is located at the start address of the redundant data storage area, and P (D21 to D24) to P (D37 to D40) are the address of the start address of the redundant data storage area + 1 (the size of each data block in the stripe). Placed in

However, the disk storage devices 12-1 to 12-12
In order to be able to recover data even if one of the -5s fails, the data must be located in a disk storage device different from the disk storage device in which the data is located. Therefore, S
The components D1 to D4 of S1 are disk storage devices 1
2-1 to 12-4, so that S
P (D1 to D4), which is redundant data of S1, is arranged in the disk storage device 12-5. Similarly, P (D5 to D8) which is redundant data of SS2 is stored in the disk storage device 12-4.
To place.

An example of the configuration when the number of disk storage devices is generalized to N + 1 will be described with reference to FIG. Here, N is the number of disk storage devices for data used for striping. The input data is arranged in the storage area of the input data in the order from the disk storage device 12-1 to the disk storage device 12- (N + 1), and the disk storage device 12- (N + 1) is followed by the disk storage device 12- (N + 1). Storage device 12-1
And is located at the next storage location.

The arrangement of the redundant data in the storage area of the redundant data is arranged in a disk storage device in which the striped data corresponding to the redundant data is not arranged.
For example, the redundant data P (DN + 1 to D2N) is arranged in the disk storage device 12-N in which the corresponding data DN + 1 to D2N are not arranged.

In FIGS. 2 and 3, there is a redundant data area continuous with the data area. The redundant data area may be anywhere on the inner peripheral side of the disk with respect to the data area, and is continuous with the data area. No need.

The processing of the RAID controller 5 of the present embodiment, including the method of setting these areas, is initialized,
The process for securing a necessary area before writing, the process for writing, the process for erasing, and the process for reading will be described separately.

First, the initial setting will be described. In the initial setting, a zone information table indicating an address range for each zone of the ZBR system is created. In this creation method, the boundaries of the zones of the disk in the disk storage device are determined, the corresponding physical address range is checked in advance, and the result is stored in the memory. As a method of determining a zone boundary, for example, there is a method of performing a binary search for a zone boundary value, but the method is not limited to this. Since the initial setting is performed before the operation of the server system, it does not affect the writing or reading performance of the server system. This zone information table is stored in the memory 9.

Next, processing for securing a necessary area before writing will be described. This will be described with reference to the flowchart of FIG.
Before the write command, the host CPU 2 of the server
The name and size of the data to be written are specified and sent to the ID controller 5. In step S101, the size of an area necessary for storing data in the disk storage device is determined from the designated size, and the size of the area is written in the memory 9.

Steps S102 to S104
Creates a data area list for arranging striped data and stores it in the memory 9. Step S102
Then, the size of the area determined in step S101 is
Using the zone information table created in the initial setting and the unused area list in the memory 9, an unused area is allocated from the outer peripheral side of the disk in the disk storage device and secured. The unused area list is a list including the start address of the unused area and the size of the area.
The unused area may exist continuously in the disk storage device or may exist in a fragmentary manner.

In step S103, an area of the size secured in step S102 is written in the data area list to be created. In step S104, it is determined whether the required size has been secured. If the required size has not been secured, the process returns to step S102. If the size has been secured, the process proceeds to step S105. The data area list is stored in the memory 9.

Steps S105 to S108
Creates a redundant data area list for arranging redundant data and stores it in the memory 9. In step S105, the size of the area required for storing the redundant data is calculated. Assuming that the redundant data is one data per stripe of the corresponding data, the size of the area required for the redundant data is calculated by dividing the size of the data area by the number of divisions of the data constituting one stripe.

In step S106, step S105
Using the zone information table created by the initial setting and the unused area list in the memory 9, the area corresponding to the size calculated in the above is stored in the disk storage where the corresponding data is not stored on the inner side of the acquired data area. By allocating an area to the device, a redundant data area is secured.

In step S107, the area of the size secured in step S106 is written in the redundant data area list to be created. In step S108, it is determined whether the required size has been secured. If the required size has not been secured, the process returns to step S106.
Go to 9.

In step S109, the file system is initialized. The contents include the logical addresses of the input data and the redundant data, and the contents of steps S101 to S1.
The mapping between information and the physical address of the area secured in step 08 is performed, and the result is stored in the memory 9 as an address mapping table. The address mapping table includes a logical address, a physical stripe number, and a physical address.

Next, the data writing process will be described with reference to the flowchart of FIG. For the sake of simplicity, it is assumed that the head address of the secured data area is physical address 0, and the address value increases toward the inner circumference. The secured data area is assumed to be continuous.

In step S201, the logical address of the data passed from the OS (Operating System) of the RAID controller 5 is converted into a physical stripe number using the address mapping table created in step S109. That is, the physical stripe number is determined from the logical address and converted.

In and after step S202, the physical address at which the individual data is to be written is calculated. Step S201
The data to be written to which address of which disk storage device in the physical stripe number obtained in the above is calculated as follows. The logical address is RA (block position) and the data division size in the stripe is SZ (bloc
k), the number of disk storage devices constituting the RAID is DK. Since the size in which data can be actually written is reduced by the size of one disk due to the redundant data, the number of disk storage devices for writing data is DK-1.
Thus, a data stripe is configured.

Since the size of the data that can be written in one stripe is SZ × (DK−1), the quotient of equation (1) determines the number of the disk storage device in the stripe number to be written. Is calculated. In equation (1),% is the remainder of the division. (RA% (SZ × (DK-1))) / SZ (1) Further, the offset position from the head address corresponding to the physical stripe number in the disk storage device is calculated based on the remainder of the equation (1). be able to.

As an example, as shown in FIG. 2, a RAID is composed of five disk storage devices, and the stripe size SZ is 1
Consider the case assuming 6 blocks. In this case, the number of disk storage devices for data storage is four, excluding the size for redundant data, so that DK-1 = 4.

When 81 is passed as the logical address from the OS, (81% (16 × 4)) / 16 = 17/16 = 1 (quotient) is obtained from the equation (1). It can be seen that, of the plurality of disk storage devices in the physical stripe number, writing to the drive at the head of the stripe plus the first drive, that is, the disk storage device at the position next to the head, is sufficient. Also,
Since the remainder of the equation is 1, the offset of the address for writing data into the disk storage device is:
It can be seen that it is one block worth.

Note that the disk storage device number at which the stripe starts depends on the physical stripe number, which is obtained by equation (2). In equation (2), SN is the physical stripe number of the data, DK is the number of disk storage devices constituting the RAID, and% is the remainder of the division. However, the physical stripe number starts from 1. The numbering of the disk storage devices is also performed in the order from 1. DK-((SN + DK-2)% DK) (2)

For example, in five disk storage devices, the disk storage device number at which the stripe whose data physical stripe number is 3 starts is 4 by substituting DK = 5 and SN = 3 into equation (2). No. 4 This result is confirmed by the fact that the physical stripe number 3, ie, the leading data D9 of SS3 is in the disk storage device 12-4 in FIG.

In step S203, step S201
Calculate the physical address to write the redundant data corresponding to the physical stripe number converted by the above. A disk storage device to which redundant data is written needs to allocate a disk storage device in which data corresponding to a physical stripe number is not arranged. More specifically, redundant data is written to a disk storage device at a position immediately before a disk storage device in which the first data in the stripe is written.

At which address of the disk storage device the redundant data corresponding to the data to be written is to be written
Assuming that the leading physical address of the redundant data area is 0 for convenience, it is calculated by equation (3). In the expression (3), INT means an integer value, which is a position offset by the quotient excluding the remainder of the division in parentheses. INT ((SN-1) / DK) (3)

Actually, assuming that D1, D2, etc. in FIGS. 2 and 3 are striped data blocks of a certain size, only the product of the result of equation (3) and the number of striped data blocks is obtained. The offset address becomes a position where redundant data is written.

For example, in five disk storage devices, redundant data having a physical stripe number of 6 is DK =
By substituting 5, SN = 6 into equation (3), 1 is obtained, and it can be seen that this is the address of the top physical address of the redundant data area + 1 (the size of each data block in the stripe). This result is confirmed by the fact that the redundant data P (D21 to D24) of the physical stripe number 6 in FIG. 2 is the first physical address of the redundant data area + 1 (the size of each data block in the stripe). All physical addresses corresponding to the data input in steps S201 to S203 are specifically determined.

In step S204, redundant data is calculated. This may be performed by using the redundant data calculation circuit 10 of hardware for the input data, or calculating the redundant data by software processing according to a RAID algorithm. In step S205, the striped data is written to the data physical address, and the redundant data corresponding to the striped data is written to the redundant data physical address. As described above, all the writing processes are performed.

In the write processing, since the position where data is written and the position where redundant data are written are distant from each other, if writing is performed sequentially for each data, many seeks occur, which is disadvantageous in terms of access speed. The redundant data is collectively stored in the memory 9 in the controller 5, and is written to the disk storage device when the redundant data is collected to some extent.

Next, processing at the time of erasing will be described. When a predetermined data erasure instruction is received from the server, the RAID controller 5 moves the contents of the data area list of the predetermined data to the unused area list, and changes the contents of the redundant data area list of the predetermined data to the unused area list. Move. In this way, the unnecessary data area and redundant data area are released, and can be used for data allocation at the time of next data writing.

Next, the reading process will be described with reference to the flowchart of FIG. Normally, only data is accessed without accessing redundant data. This is because redundant data is usually unnecessary. The name and size of the data to be read are instructed from the host CPU 2 of the server to the RAID controller 5.

Step S301 is the same as step S201, and step S302 is the same as step S202. That is, the O of the RAID controller 5
The logical address of the designated data passed from S is converted into a physical stripe number, and the physical address from which data is read is calculated. In step S303, step S3
01, data of a required size is read from the disk storage device from the physical head address calculated in S302.

During normal reading, only the reading of the data area is performed, so that the seek operation to the inner peripheral side with the redundant data is not performed. In the ZBR method, the data transfer speed is higher at the outer periphery, and the data transfer speed is higher because the data to be stored is arranged on the outer periphery of the disk storage device. Also, since the data is arranged collectively, the data transfer speed is increased as much as the redundant data is not included. Faster data transfer speed means, conversely, shorter data transfer time.

Further, as shown in FIG. 7A, the conventional seek range is the data and redundant data area 20, but as shown in FIG. 7B, the data area 21 and the redundant data area 22 are divided. If they are separated, the seek range only needs to be the data area 21, and the seek range is narrowed, so that the access time can be reduced.

It is necessary to read the redundant data from RA
This is a case where one of the disk storage devices constituting the ID has failed. Even in this case, the data written in the normal disk storage device and the redundant data are transferred to step S2.
02 and the same method as in step S203, the addresses of the stored data and the redundant data are specified and read out, so that the contents written in the failed disk storage device can be recovered. That is,
Even if one unit fails, the original function of RAID that data can be recovered is not lost.

According to this algorithm, if one disk fails, a full seek may occur, but the probability of failure of one of the disk storage devices constituting the RAID is determined by the probability that all disk storage devices will fail. Since the device is lower than the case where the device is normal, there is little practical problem. For example, with current hard disk storage devices, one MTBF
(Mean Time Between Failure) is one million hours, so in a RAID using ten of them, the MTBF until one of them fails is 100,000 hours, and access to redundant data is rare. This does not occur, and is usually only data access.

Further, in the case of a completely duplicated server having completely the same data, when one disk storage device of one server fails, another server takes over the operation. Therefore, there is no need to consider a decrease in service quality at the time of failure.

As an embodiment, the RAID controller 5
A function of calculating redundant data only by software processing of the host CPU 2 and determining an arrangement position of input data and redundant data to be originally stored in the disk storage device. A form having a function of only distributing data to the devices 12-1 to 12- (N + 1) is also possible. Also,
A form in which some of these functions are realized by the firmware of the disk storage devices 12-1 to 12- (N + 1) is also possible.

The disk of the disk storage device may be a hard disk or an optical disk as long as the disk is formatted by the ZBR system, or a disk which can be formatted by the ZBR system. It may be in a form.

[0065]

According to the disk array control method and the disk array device of the present invention, input data to be originally stored and redundant data are divided, and the input data is stored from the outer peripheral side of the disk of the ZBR type disk storage device. By storing redundant data that is used only when the disk storage device has failed on the inner side of the disk from the input data,
Data transfer time during normal reading can be reduced. Therefore, VOD (Video On D
With an information distribution server such as emand), it is possible to reduce the data transfer time at the time of reading and the access time. In addition, the same effect can be obtained for a database server that is generally read frequently.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a configuration example of an information distribution server.

FIG. 2 is a diagram illustrating an example of a data arrangement in a case where the disk is configured by five disk storage devices.

FIG. 3 is a diagram showing an example of a data arrangement in the case of being configured with N + 1 disk storage devices.

FIG. 4 is a flowchart of a process for securing a necessary area of the RAID controller.

FIG. 5 is a flowchart of a write process performed by a RAID controller.

FIG. 6 is a flowchart of a process at the time of reading by the RAID controller.

7A and 7B are diagrams showing a seek range at the time of reading, FIG. 7A is a diagram showing a conventional seek range, and FIG. 7B is a diagram showing a seek range of the present embodiment.

FIG. 8 is a diagram showing a method of allocating data in the RAID-3 system.

FIG. 9 is a diagram showing a method of allocating data in the RAID-5 system.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Host bus, 2 ... Host CPU, 3 ... Memory, 4 ...
Network adapter, 5 ... RAID controller, 6
... BIU, 7 ... control CPU bus, 8 ... control CPU, 9 ... memory, 10 ... redundant data calculation circuit,
11: Disk controller unit, 12-1 to 12- (N
+1) Disk storage device, 13 Disk bus, 20
... data and redundant data area, 21 ... data area, 22
... redundant data area, SS1, SS2, SS3, SS4,
SS5, SS6, SS7 ... Striped data set

Claims (6)

[Claims] 1. A disk array control method in which redundant data is added to input data and the input data and the redundant data are separately stored in a plurality of disk storage devices having a zone bit recording type disk. A disk array control method, wherein a storage area of data on the disk is arranged on an inner peripheral side of the disk with respect to a storage area of the input data. 2. The disk array control method according to claim 1, wherein a storage area of the input data is arranged from an outer peripheral side in a usable area of the disk. 3. A method of arranging the input data in a storage area of the input data, wherein the number of divisions of the input data is N, the total number of the disk storage devices is N + 1, and the number of the disk storage device is 1 to 1 N
When +1 is set, the disk storage device of the number 1 is arranged in order from the disk storage device of the number 1 to the disk storage device of the number N + 1. The method of arranging the redundant data in the storage area of the redundant data is characterized in that the redundant data is arranged in a disk storage device in which the divided input data corresponding to the redundant data is not arranged. 2. The disk array control method according to claim 1, wherein: 4. A disk array device in which redundant data is added to input data and the input data and redundant data are separately stored in a plurality of disk storage devices having a zone bit recording type disk. The disk array device according to claim 1, wherein a storage area of said disk is arranged on an inner peripheral side of said disk with respect to a storage area of said input data. 5. The disk array device according to claim 4, wherein a storage area for the input data is arranged from an outer peripheral side in a usable area of the disk. 6. An arrangement configuration of the input data in a storage area of the input data, wherein the number of divisions of the input data is N, the total number of the disk storage devices is N + 1, and the number of the disk storage device is 1 to 1. N
When +1 is set, the disk storage device of the number 1 is arranged in order from the disk storage device of the number 1 to the disk storage device of the number N + 1. The arrangement of the redundant data in the storage area of the redundant data is such that the divided input data corresponding to the redundant data is arranged in a disk storage device in which the divided input data is not arranged. The disk array device according to claim 4, wherein: