JP2774728B2 - Disk array control method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a disk array control system for high-speed data transfer in a disk array controller.

[0002]

2. Description of the Related Art A disk array device is connected to a plurality of disk devices and can read and write data in parallel, thereby realizing a higher processing speed than a single disk device. .

The disk array device is composed of a plurality of disk devices and a disk array control device for controlling the disk devices. The computer connected to the disk device (upper computer) looks like a single disk device. Therefore, it is necessary to convert a command issued from a host computer into a command for each disk device in the disk array control device. Such a data transfer control method is disclosed in, for example, Japanese Patent Application Laid-Open No. 62-13132.
No. 1 publication. Hereinafter, this will be described with reference to FIG.

In FIG. 16, means 13 for performing the above-mentioned command conversion and distributing the command to each disk device is provided.
Means for issuing a command to each disk device and detecting completion of read and write data transfer.
, A command that receives a command from a host computer, and notifies the end of the command, a unit 17 that controls data transfer between the host computer and each disk device, and a unit 16 that controls these units. A single command from the host computer is assumed to be a single command to each of the multiple disk devices, and multiple disk devices are operated in parallel in the same way, realizing high-speed data transfer. Is what you do.

Further, the disk array device connects a large number of disk devices, and the failure rate of the disk increases in proportion to the number of disk devices, which is a problem in terms of high reliability. Therefore, a group is composed of several data disk devices and one parity disk device. Further, data is divided into data units called stripes, and all stripes starting from the same logical address of all disk devices in the group constitute a group called parity group. Parity information is generated by calculating exclusive OR of all data stripes in the parity group, and this is stored in the parity disk device as a parity stripe. In this way, by giving data redundancy instead of simply distributing data to a plurality of disk devices, even if one disk device fails, the failed disk device can be obtained from the parity information of the parity disk device. Conventionally, a high-reliability scheme configured to be able to restore the above data has been discussed.

Japanese Patent Application Laid-Open No. Sho 64-53224 discloses a method of reading and controlling file data over a plurality of continuous fixed-length blocks at high speed in a disk device. According to this method, when reading file data, it is checked whether there is a continuous block in the block in which the file data is stored,
When there are continuous blocks, these blocks are collectively read to enable high-speed data transfer with reduced rotation waiting time of the disk device.

[0007]

The host computer accesses the disk array device to a continuous area on a logical address, but divides the command into a plurality of commands.
In some cases, these are continuously transferred and a request is issued. That is, the OS (operating system) of the host computer may handle data by dividing the data into units called pages of a certain fixed size. In this case, a long data access request is issued to the disk array device as a plurality of commands. Also, the program running on the host computer is
When performing a process such as generating a number of data files in the same procedure, the same operation occurs because these data files are often arranged in a continuous area.

However, the above-mentioned prior art does not take into consideration the processing at the time of accessing the disk array device to such a continuous area a plurality of times (hereinafter, referred to as a plurality of sequential accesses). According to this conventional technique, even if the above-mentioned sequential continuous access request is issued to the disk array device, the disk array device processes them as a single command. There is a problem in that a disk rotation wait occurs for each command, and the processing speed is greatly reduced.

In the above prior art, it is necessary to generate parity information as described above. Prior to data writing, old data and old parity information already written in an area to be written are read, and these are read. New parity information is generated by calculating an exclusive OR with the new data to be written. According to this conventional technique, at the time of the above-described sequential write access, the old data read processing for generating the parity information is performed by 1
Since each command is executed for each command and two disk rotation waits are required for one write command,
There is a problem that the processing speed is reduced in addition to the rotation waiting for each command.

Further, in the above-mentioned conventional technology, since continuous data on a logical address is distributed to a plurality of disk devices from a host computer, data is not arranged continuously on the disk devices. Therefore, as a function normally included in the disk device, when a read command is issued, not only the data but also the continuous data following the data is pre-read, and when the next command accesses this area, the operation of the mechanism unit such as seek and rotation waiting. Although there is no waiting time and there is a function (prefetch function) for transferring the prefetched data immediately, according to the above-described conventional technology, since the data is not arranged continuously on the disk device, the prefetch function of this disk device is provided. Does not work, and the operation of the mechanism unit waits for each access, so that the processing speed at the time of a plurality of sequential accesses is lower than that of a single disk device.

Further, in the above-described method of performing batch reading of continuous blocks in a disk device, it is assumed that a file is stored in a single disk device, and a file is divided into a plurality of stripes as in a disk array device. No consideration is given to the case of division and recording in a distributed manner on a plurality of disk devices. Also, only the case where the file is divided into fixed lengths is shown, and no consideration is given to the case where the OS of the host computer divides the file with a variable length block size. Further, only the read operation is shown, and as described above, no consideration is given to speeding up the write operation, which particularly reduces the processing speed, in the disk array device.

An object of the present invention is to provide a disk array control method which solves such a problem and does not cause a disk rotation wait at the time of a plurality of sequential accesses divided into blocks of arbitrary length.

It is another object of the present invention to provide a disk array control system which does not require a read operation for generating parity at the time of a plurality of sequential accesses for performing the write operation.

Still another object of the present invention is to provide a data transfer function for a read operation as well as a write operation without an operation waiting time of a mechanical unit of the disk device by a function replacing the prefetch function of the disk device. To provide a disk array control method for performing the following.

[0015]

In order to achieve the above object, the present invention provides a means for receiving a command from a host computer, notifying the end of the command, and interpreting a command issued from the host computer. Means for converting the command to a command to each disk device, activating a disk command to each disk device, and detecting completion of read or write data transfer by the disk command,
In a control method of a disk array device including means for controlling data transfer between the host computer and each of the disk devices, and means for controlling each of these units, a command continuously issued by the host computer is provided. Means for temporarily storing them in a queue and managing them; and when a plurality of commands stored in the queue are commands requesting sequential access, these are grouped and regarded as one command. Means for converting a command into a disk command for each disk device, and storing and managing information on the grouped command and conversion information for converting the grouped command into a command for each disk device. Means are provided.

In order to achieve the other object, the present invention further provides a first control means for controlling so that continuous data following data handled by a command issued by a host computer is read in advance from a disk device. And data storage means for storing the prefetched data inside the disk array control device or each disk device, and second control means for input / output control of the prefetched data to the data storage means. Provide.

[0017]

The command queue management means successively stores a plurality of sequential access commands issued by the host computer. The command interpreting means interprets these, recognizes that the request is a sequential access request, and notifies the command grouping means of these commands. The command grouping means considers these command groups as one command and groups them, and stores and stores the correspondence in the conversion information management means. Further, the command grouping means converts the one grouped command into a disk command for each disk device, and stores the correspondence of the conversion in the conversion information management means.

By the operation of each section described above, a plurality of sequential access commands are converted into one sequential access command inside the disk array controller, and each data can be handled continuously and collectively. Therefore, there is no disk rotation waiting between commands, and the data transfer speed can be increased.

Further, at the time of writing, the command grouping means operates as follows. First, as described above, a plurality of sequential write access commands are grouped. Then, a data area for writing all data stripes in the parity stripe is cut out. In the data area other than the above data area, data read for parity generation is performed as in the past, parity is generated, and then write operation is performed. In the data area, parity information is directly generated from the data stripe and only write operation is performed. Perform Since the data is grouped as described above and the data is handled collectively and continuously, the data area increases, and the write operation of most data does not require a read operation for parity generation. Achieve higher speed.

If the command interpreting means interprets the command issued by the host computer and detects that the disk device storing the continuous data area in the data area handled by the command is not accessed, the data prefetching is performed. The control means receives this and issues a read request to the disk device. This disk device does not have the data requested by the higher-level computer, but responds to the read request issued by the data pre-reading control means and stores the next data following the data requested by the higher-level computer in the pre-read data storage means. And send it out. At this time, the means for controlling the data transfer inputs the read data to the prefetch data storage means. Then, when the command issued by the upper-level computer ends, and the upper-level computer next requests sequential data read that is continuous with the data of the previous access, the data pre-reading control unit uses the data input to the pre-reading data storage unit first. Judge that it is requested, and immediately send this data to the host computer. As described above, since the disk device is not actually accessed, a high-speed data transfer at the time of reading is realized.

Also, at the time of writing, the read processing is performed for parity generation as described above. At this time, the data pre-reading control means operates to perform the pre-read operation in the same manner as described above. When the current write command is completed and the next sequential write command is issued from the host computer, the old data required for parity generation is stored in the prefetch data storage means, so that the read processing is not performed. Parity data can be generated and data can be written immediately,
High-speed data transfer at the time of writing is realized.

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS First, a disk array device using the present invention will be described with reference to the drawings. FIG. 2 is a block diagram showing the configuration of the disk array device, 1 is a host computer to which the disk array device 2 is connected, 3 is a disk array control device that controls the disk devices 9a1 to 9en arranged on the array, Reference numeral 4 denotes an upper interface (hereinafter, upper I / F) for connecting the disk array device 2 and the upper computer 1
5 is a data control means for controlling the flow of data in the disk array device 2, and 6 is a command control for interpreting a command sent by the host computer 1 and converting it into commands for the respective disk devices 9a1 to 9en. 7 is a data buffer, and 8a to 8e are disk interfaces (hereinafter referred to as disk I) for connecting the disk array control device 3 and disk devices arranged on the array.
/ F), 10 is one disk drive 9a1-9e
This is a parity generation means for generating parity information provided to prevent the data from being lost even if any of n fails.

In FIG. 1, the disk array controller 3
Then, the read command or the write command from the host computer 1 received by the host I / F 4 is transmitted to the command control unit 6.
Is forwarded to The command control means 6 interprets the received command, selects a disk device to be accessed, and generates a command for the disk device.

Here, in order to avoid confusion in the following description,
The command sent by the host computer 1 is called a host command, and the command unique to the disk device sent to each of the disk devices 9a1 to 9en is called a disk command.

The disk command generated by the command control means 6 is transferred to the target disk device via a disk I / F to which the target disk device is connected, and causes a read or write process to be executed. .
At this time, it is also possible to select a plurality of disk devices and operate them simultaneously. Therefore, higher performance operation can be expected as compared with a single disk device.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a specific configuration of the command control means 6 in FIG. 2 using one embodiment of a disk array control system according to the present invention. Reference numeral 11 denotes a host command transmitted from a host computer 1 via a host I / F 4. , A host command input / output means for reporting the end of the command, 12 a host command queuing means for temporarily storing and managing the received host command in a queue, and 13 a disk command for interpreting the received host command and interpreting the received host command. , A command clustering unit that integrates a plurality of host commands to generate a command cluster, and 15 issues a generated disk command to each disk device, and also issues a command from each disk device. Start disk command to receive termination
Ending means, 16 is a main control means for managing and controlling the entire command control means 6, 17 is a data transfer timing control means for instructing the data control means 5 (FIG. 2) of an appropriate transfer timing, and 18 is a command clustering means 1
4 is a disk command management table for managing the correspondence between the command cluster generated in Step 4 and the original disk command.

Next, the operation of the disk array device 2 will be described. In FIG. 2, from the host computer 1, the disk array device 2 appears as one large-capacity disk device. Now, it is assumed that a command that is continuous data access but is divided into a plurality of times has been transmitted from the host computer 1 to the disk array device 2. Such an access request occurs in the following cases.

First, there is a case due to the restriction of the OS. If you are working with long sequential files,
If S does not handle the data length of this file collectively,
The command is issued a plurality of times by dividing the data.

Second, it may occur depending on the characteristics of the application. If an application regularly creates or reads multiple files, it is likely that they will be located on consecutive logical addresses.
Multiple sequential accesses are issued.

The operation when such multiple sequential accesses are issued from the host computer 1 to the disk array device 2 will be described with reference to FIG.

FIG. 3 shows a logical address map of the disk array device 2 as viewed from the host computer 1 as a host. In FIG. 3, the host computer 1 has a command A (A) of a logical address AhA and a data length LA. Hereinafter, simply A
Logical address AhB, data length LB command B (hereinafter simply referred to as B), logical address AhC, data length LC command C (hereinafter simply referred to as C), logical address AhD and data length LD command D Assume that four host commands (hereinafter simply referred to as D) are issued to the disk array device 2 in succession. The data handled by these host commands A, B, C, and D are arranged continuously on the address map as shown in FIG.

First, the host command A is sent to the upper I / F 4
Via (FIG. 2), it is supplied to the command input / output means 11 in the command control means 6 shown in FIG. 1, and the command input / output means 11 transfers this to the host command queuing means 12. If the internal command queue is not full, the host command queuing means 12 receives the host command A and stores it at the end of the command queue. One host command is extracted from the head of the queue and sent to the host command interpreting means 13.

Now, the host command queuing means 1
If the host command A is stored at the head of the command queue in Step 2 and the host command interpreting means 13 is not being executed, the host command A is read from the command queue and transferred to the host command interpreting means 13, The command is interpreted and converted to a disk command. At this time, the host command queuing unit 12 notifies the host computer 1 via the host command input / output unit 11 that the host I / F 4 can receive the next host command. At the same time,
F4 is ready to receive the next host command,
If the host computer 1 can send the next host command, the host computer 1 sends a host command B, followed by a host command C and D each time there is a notification from the host command queuing unit 12, and sends a host command. I do. The host command queuing unit 12 sequentially stores these host commands in a command queue, and the host command interpreting unit 13 stores these host commands B, C, D
In the same way as the host command A, the command is interpreted and converted into a disk command.

When a plurality of host commands are successively transmitted from the host computer 1 and stored in the command queue, the command clustering unit 14 immediately targets the disk command converted by the host command interpreting unit 13. Instead of sending it to the disk device to be executed,
To be stored. If these consecutively received host commands A to D are not access to continuous logical addresses of continuous data, the command clustering means 14 executes a disk command command so as to execute the commands sequentially in an appropriate order. The disk commands are taken out from the management table 18 and sent to the disk command start / end unit 15 to issue these disk commands to each disk device.

On the other hand, if these host commands A to D are accesses to a continuous area as shown in FIG. 3, the command clustering means 14 groups these host commands A to D. That is, as shown in FIG. 4, the host command A sent from the host computer 1,
B, C, and D are logical address AhA and data length (LA +
This is equivalently converted into one host command of (LB + LC + LD) (this is called command clustering).

Next, the clustered host commands are sent to the command interpreting means 13, converted into disk commands, and issued to the target disk device by the disk command starting / ending means 15.

FIG. 5 is a diagram showing mapping of each host command to a logical address of each disk device. In addition,
Here, in order to simplify the description, five disk devices a to e are connected in parallel (that is, in FIG. 2, five disk I / Fs 8a to 8e are provided in the disk array control device 3). It is assumed that one disk device is connected to each of the disk devices.) Of course, the number of disk devices connected in parallel is arbitrary, and
The number of disk devices connected to the same disk I / F is also arbitrary.

Of these five disk devices a to e, disk devices a to d are for reading and writing data, and disk device e is for reading and writing parity data. .

Data handled by one host command is divided into unit blocks called stripes shown in FIG. 5 and recorded on a disk. Here, the host command A
Is divided into three stripes A1, A2 and A3, and these stripes A1 to A3 are recorded on the disk devices a, b and c. Data handled by the host commands B, C, and D are similarly divided into two, five, and one stripes, respectively, and are recorded on the disk devices a to e as shown in FIG.

Next, the disk array control system according to the present invention used here is divided into a read operation and a write operation, and the host command shown in FIG. 3 will be described in detail.
First, a read operation will be described. For comparison, first, a case where command clustering is not performed unlike the conventional method will be described.

As described above, the read data of the host command A is divided into stripes A1, A2, and A3 and recorded on the disk devices a, b, and c. Commands Aa, Ab, A
The read by c is performed. Similarly, for the host command B, the stripes B1 and B2 are obtained from the disk devices d and a by the disk commands Bd and Ba, and for the host command C, the stripes C1 and C5 are obtained from the disk device b and the disk command Cb. C3 and C4 are C from disk devices c, d and a, respectively.
According to c, Cd, and Ca, for the host command D,
The stripe D1 is read from the disk device C by the disk command Dc.

Referring to FIG. 6, when the host commands A, B, C and D are successively supplied from the host computer 1 to the command conversion means 6 via the host I / F 4,
The command conversion means 6 shown in FIG. 1 temporarily stores these in the host command queuing means 12. First, the host command interpreting means 13 first reads the disk devices a, b,
host command A to disk command A
a, Ab, Ac are issued, and the respective disk devices a, b,
c, these disk commands are executed almost simultaneously.
When each of the disk devices a, b, and c receives the disk command, it interprets the disk command, and after a seek operation and disk rotation wait, detects a target sector on the disk and reads the target data. Thus, the data A1, A2, and A3 read from the disk devices a, b, and c are recombined as read data of the host command A, sent to the host computer 1, and the execution of the host command A ends.

At this time, the command conversion of the host command B has already been executed, and the disk command B is transferred to the empty disk device d almost simultaneously with the host command A.
d can be issued, but the disk device a is executing the host command A.
Cannot be issued, and it is necessary to wait for this execution to end. After confirming the end, a disk command B is sent to the disk device a.
a is issued, but as shown in FIG.
Although the data areas of the stripes A1 and B2 are continuous, the head address of the stripe B2 passes during an overhead such as a command interpretation time, so that there is no need to wait for rotation around the disk 1. must not. The same is true for the host commands C and D. In general, when executing a host command for performing a continuous n number of sequential reads, the disk rotation waits up to n times, which lowers the processing speed.

Next, a read operation according to the disk array control method of the present invention used in FIG. 1 will be described. In FIG. 1, the host commands A, B, C, and D are once grouped (command clustering) by the command clustering unit 14 in the command control unit 6 as described above, and as shown in FIG. After pretending to be equivalent to a command, the command is converted into a disk command and a disk command is issued to each disk device. This is shown in FIG.

In the figure, host commands A, B, C, and D are received from the host computer 1, and the command control means 6 interprets them and determines that the access is a sequential access to a continuous area. , The command clustering means 14 performs command clustering. Then, the host command interpreting means 13 interprets the stripe area for each of the disk devices a to d to be read by this one command clustered host command, and, based on the result, the disk command CCa for each of the disk devices a to d. , CCb, CCc, and CCd are generated and issued to the respective disk devices a to d. here,
The disk command CCa indicates the stripes A1, B2 and C4 on the disk device 9a, the disk command CCb indicates the stripes A2, C1 and C5 on the disk device 9b, and the disk command CCc indicates the stripes A3, C2 and D1 on the disk device 9c. , Disk command CCd is a disk command for sequentially and successively reading the stripes B1 and C3 on the disk device 9d, respectively.

These disk commands CCa to CCd are simultaneously issued to the respective disk devices a to d, and after seeking and rotation waiting, read respective target data. After these data are once transferred to the data buffer 7 (FIG. 2),
On the basis of the management information stored in the disk command management table 18 (FIG. 1), the data transfer timing means 17 transmits the data from the data buffer 7 to the stripes A1, A2,.
.., C5, and D1 are read in the order of the data and the host computer 1
And the host commands A, B, C, and D are sequentially terminated.

By executing the command clustering in this way, when n sequential access commands from the host computer 1 are issued to the disk array device 2, the number of rotation waits is only one, and the conventional method is used. In comparison, the number of disk rotation waits can be reduced by (n-1) times. Normally, the rotation waiting time is much longer than the data transfer time, so that the performance improvement effect by reducing the number of rotation waiting times is very large.

Next, this effect will be described in more detail with reference to specific examples. Now, it is assumed that the average seek time of each disk device is 15 milliseconds, the average rotation wait time is 8 milliseconds, the rotation wait time of one rotation of the disk is 16 milliseconds, and the data transfer speed of the disk device is 3 Mbytes / second. . Also,
The number of disk devices used for the read operation is four. At this time, an access request is issued from the host computer 1 to the 64 Kbyte continuous area in four 16 Kbyte host commands, and each 16 Kbyte access is performed by four units. It is assumed that data stored in the disk device is read evenly distributed. Furthermore, the overhead time for command interpretation and the like is very small and can be ignored.

Under these conditions, according to the conventional method, the processing time T 'is T' = (average seek time + average rotation wait time) + (16 Kbytes data transfer time / 4) .times.4.
+ (One rotation waiting time) × 3 = 76.3 (milliseconds). On the other hand, according to the disk array control method of this embodiment, the processing time T is T = (average seek time + average rotation waiting time) + (64 Kbyte data transfer time / 4) =
28.3 (milliseconds), which is about three times faster than the conventional method. The greater the number of sequential access commands from the host computer 1 to the continuous area, the greater the effect of this embodiment.

Further, according to this embodiment, FIGS.
As is clear from comparison with the above, it can be seen that the occupation time of the disk device is shorter than that of the conventional method. If the next host command is received during the execution of these host commands, the conventional method will wait a long time until the received command can be executed.
According to this embodiment, the occupation time of the disk device is short, so that the next command can be executed with a short waiting time, the processing throughput is increased as a whole, and the average of the commands as viewed from the host computer is increased. This has the effect of shortening the response time.

Next, the operation at the time of writing will be described.
As in the case of reading, the host command shown in FIG. 3 is mapped to each disk device and written as shown in FIG. 5. At the time of writing, it is necessary to generate the above-described parity data. And the control method is different. A method for creating the parity data will be described with reference to FIG.

In the disk array device of this embodiment, 5
One of the disk devices is assigned as a parity disk device. As described above, the disk device e in FIG. 5 is the parity disk device. Parity is created in a group called a parity group in units of stripes. One parity group is formed by stripes starting from the same logical address of each of the disk devices a to e, and the exclusive OR of these data is used as parity data. In FIG. 5, for example, stripe A1,
A2, A3, B1, and P1 form one parity group, and a stripe P1 of parity data is generated from an exclusive OR of the stripes A1, A2, A3, and B1.

Here, for comparison, a conventional method will be described first with reference to FIG. The host computer 1 as a host sends a write request for continuous data to four host commands A, B,
The commands are issued as C and D, and the command control means 6 of the disk array controller 3 receives and interprets them, converts them into disk commands, and sequentially activates the disk devices, similarly to the read operation. Here, the difference from the lead time is
The point is that, before writing, a read operation of reading old data written on the disk is performed, and an exclusive OR of these data and new data to be written is generated to generate new parity data.

Specifically, the old data A1 ', A of the stripes A1, A2, A3 corresponding to the host command A
2 ′, A3 ′ and the old parity stripe P1 ′, and then read the old data and the new data stripe A
A new parity data stripe P1 is created by taking an exclusive OR of 1, A2, and A3. Then, after waiting for one rotation of the disk, the new data stripes A1, A2, A3 and the new parity data stripe P1 are written to complete a series of write operations. Such processing is similarly performed for B, C, and D.

According to this, between the read operation and the write operation for generating the parity and between the execution of the host command and the execution of the host command, as in the case of the read operation, a rotation wait occurs and the processing speed is reduced. I do.

Next, the first write system in this embodiment will be described. As in the case of reading, upon receiving the host commands A, B, C, and D from the host computer 1, the command control means 6 interprets them, and if it determines that these are sequential accesses to a continuous area, the host command A , B, C, and D are grouped (command clustering) by the command clustering unit 14 in the command control unit 6, and as shown in FIG. , Read disk commands CCa1, CCb1, CCc1, CCd1, CCe to the respective disk devices a to d
1 are generated and issued at the same time. Disk command CCa
1 is the old data stripe A1 ', B of the disk device a
2 ′, C4 ′ and the disk command CCb1 are the old data stripes A2 ′, C1 ′, C5 ′ of the disk device b.
The disk command CCc1 is the old data stripe A3 ', C2', D1 'of the disk device c, the disk command CCd1 is the old data stripe B1', C3 'of the disk device d, and the disk command CCe1 is the old data stripe of the disk device e. Parity stripes P1 ', P2', P
Lead 3 'and P4' respectively. Disk devices a to e
Thereafter, after the seek operation and the rotation wait, the respective target data are read. Then, the old data and the new data A1, A2, A3, B1, B2, C1, C2, C3, C
4, C5, D1 and new parity data P1, P2, P
3. Generate P4. Thereafter, the new data A1 to D1 and the new parity data P1 to P4 are transferred to the read disk commands CCa1, CCb1, CCc1, CCd1, CCe1.
Using the same disk commands CCa2 to CCe2 as write disk commands, writing is performed in the same manner as for reading. Thus, the write operation is completed.

At this time, if the generation time of the parity data is sufficiently smaller than the one round rotation waiting time of the disk, the rotation during the halfway is only once between the read and the write. 2n-2) times, which is very effective in terms of speeding up.

Next, the second write method of this embodiment will be described. For this purpose, the contrivance shown in FIG. 9 is used. That is, when the host command logical address mapping is performed as shown in FIG.
Since one column writes the entire data stripe of the parity group, they can generate parity data without reading old data. Therefore, these two
As for the columns, as shown in FIG. 9, all the data disk devices a, b, c, d and the parity disk device e do not read data, but can only write (disk commands CCa1 to CCe1).

In the third column from the top in FIG.
4 ', C5', and D1 'are disc commands CCa2 and C
After reading with Cb2, CCc2 and CCe2 and generating parity with these and new data, the write disk command CC
Write new data at a3, CCb3, CCc3, and CCe3.

According to the second method, the number of rotation waits increases one to three times as compared with the first method. However, when handling data that requires a data transfer time that is sufficiently longer than the rotation waiting time, the read operation time for parity generation can be short, and the speed can be much higher than in the first method. In addition, compared with the conventional method, the rotation waiting (2
n-5) times.

As in the case of the read operation, the effect of the write operation will be specifically described with a specific example. here,
It is assumed that the host computer 1 issues a write access request to the 64-Kbyte continuous area in four 16-Kbyte commands divided into four, and the data of each 16-Kbyte command is equally distributed to four disk devices for writing. Also,
The stripe size is 4 Kbytes, and the overhead time for command interpretation and the like is very small and can be ignored.

Therefore, according to the conventional method, the processing time T ′
Is, at worst, every time after writing, T '=
(Average seek time + Average rotation wait time) + (16 Kbyte data transfer time / 4) × 4 × 2 + (One round rotation wait time) × (Total number of accesses−1) = 145.7 (milliseconds) . On the other hand, according to the first method of this embodiment, the processing time T1 is T1 = (average seek time + average rotation waiting time) + (64 Kbyte data transfer time / 4) × 2 +
(One rotation waiting time) = 49.6 (milliseconds), which is about three times faster than the conventional method.

When an access request is issued from the host computer 1 to the 4-Mbyte continuous area divided into four 1-Mbyte commands, the processing time T 'according to the conventional method is taken.
Is, at worst, every time after writing, T '=
(Average seek time + Average rotation waiting time) + (1 Mbyte data transfer time ÷ 4) × 4 × 2 + (One round rotation waiting time) × (Total number of accesses−1) = 802 (milliseconds). On the other hand, according to the first method in this embodiment, the processing time T1 is T1 = (average seek time + average rotation waiting time) + (4 MB data transfer time / 4) × 2 +
(One-rotation waiting time) = 706 (milliseconds), which is only about 1.1 times faster than the conventional method.

On the other hand, according to the second method of this embodiment, assuming that the processing time T2 has a maximum of five rotation waits and the area for performing the write operation after reading is 16 Kbytes, = (Average seek time + average rotation waiting time) + ((4
Data transfer time of (M bytes-16 K bytes) ÷ 4) +
(16 Kbyte data transfer time) × 2 + (one rotation wait time) × 4 = 438 (milliseconds), which is about twice as fast as the conventional method, and the effect is extremely large. .

When data is written or read to or from the disk devices 9a to 9e after the above read or write operation, the disk devices 9a to 9e in FIG. 1 to the disk command start / end unit 15 of the command control unit 6 shown in FIG. 1, and the disk command start / end unit 15 notifies this to the command clustering unit 14. When the above operation is a read operation, the data transfer timing means 17
The data corresponding to the previous host command stored in the data buffer 7 (FIG. 2) is recombined, and
(FIG. 2). Then, the command clustering unit 14 deletes the management information corresponding to these commands from the disk command management table 18, and notifies the host command queuing unit 12 of the deletion to delete the corresponding host command from the command queue. ,
At the same time, the host command input / output means 11 is notified. Upon receiving this notification, the host command input / output unit 11 notifies the host computer 1 of the command end via the host I / F 4 and ends a series of processing.

As described above, the effects of this embodiment can be summarized as follows: (1) At the time of sequential reading, a maximum of (n-1) rotation waits can be reduced. Thus, for example, in the case of 64K bytes of data read in which 16K bits of host commands are sequentially issued four times, the total processing time is about three times faster than in the conventional method.

(2) At the time of sequential writing of small data, according to the first method, the maximum (2n-2)
The number of rotation waits can be reduced. Thereby, for example, 16K
In the case of a 64 Kbyte data write in which byte host commands are sequentially issued four times, the total processing time is about three times faster than the conventional method.

(3) At the time of sequential writing of large data, according to the second method, the maximum (2n-5)
It is possible to reduce the number of rotation waits, and it is also possible to eliminate most of the read operation of old data for parity generation. Thus, for example, in the case of a 4 Mbyte data write in which 1 Mbyte host commands are sequentially issued four times, the total processing time is at least about twice as fast as the conventional method.

FIG. 10 is a block diagram showing the internal configuration of the command control means 6 of the disk array device 2 of FIG. 2 using the disk array control method according to the present invention. The command control means 6 shown in FIG. Control means 19
Is provided. In the following description, it is assumed that the command clustering means 14 and the host command queuing means 12 described in FIG. 1 are not provided, but these may be used as shown in FIG.

In FIG. 10, prefetch control means 1
Numeral 9 is means for judging whether or not a prefetch described later hits, and if hit, controlling to read data from a prefetch buffer described later.
Other components are the same as the command control means 6 shown in FIG.

FIG. 11 is a block diagram showing the main part of the disk array device 2 shown in FIG. 2 using the command control means 6 shown in FIG.
Reference numeral 9 denotes a prefetch control unit in the command control unit 6 shown in FIG. 10, and 20a to 20e denote disk devices 9a and 9
b is a prefetch buffer for inputting and storing data stored thereon in accordance with a command from the prefetch control means 19, and portions corresponding to those in FIG. 2 are denoted by the same reference numerals.

Next, the operation of this embodiment will be described. It is assumed that the host computer 1 reads data of a continuous area on the logical address map of the disk array device 2 by a plurality of host commands. Think. When a plurality of such sequential accesses occur, it is the same as in the first embodiment described above. In such a case, assuming that one disk device is connected to the host computer 1, if the data is stored in a prefetch data storage area called a prefetch buffer of the disk device, the data is stored in this storage area. By reading data, there is no delay depending on the mechanism of the disk device, such as seeking and waiting for rotation, and since the data is read from an electrical storage area such as a memory, it can be accessed at a very high speed. . Therefore, when a read request is issued from a host computer, usually, in the disk device, prefetch control is performed to prefetch not only the data but also data following the data and store the data in a prefetch buffer.

However, in the disk array device 2, since a plurality of disk devices are arranged in parallel and data is recorded on different disk devices for each stripe, even if the logical addresses viewed from the host side are continuous, each disk device is Data is non-contiguous on the logical address of the device. For this reason, the prefetch function of the disk device cannot be expected except for a large data access exceeding the data size of the parity group. The second embodiment enables a high-speed technique to compensate for the disadvantage of the disk array device.

Next, the operation of this embodiment will be described. Here, as an example, it is assumed that two sequential access requests A and B have been issued from the host computer 1. And
As shown in FIG. 12, A1 and A2 are the stripes of the disk devices a and b accessed by the host command A, B1 and A2.
B2 is the stripe of the disk devices c and d accessed by the host command B, and P is the parity stripe of the disk device e of the parity group including A1, A2, B1 and B2.

First, the write operation will be described. As in the first embodiment, for comparison, the conventional method is first used in FIG.
This will be described with reference to FIG. First, the host command A is executed. That is, first, the old data A1 ', A2', P 'are read from the disk devices a, b, e. Then, new parity data P is generated by calculating an exclusive logical circle of these and new data, new data A1 and A2 are written to disk devices a and b, and new parity data P are written to disk device e, respectively. The command end is reported to the computer 1, and the host command A ends. The host command B is executed in the same manner, the end of the command B is reported to the host computer 1, and these two sequential write commands end.

Next, a write operation according to the method of this embodiment will be described with reference to FIGS. 2, 10, 11, and 14. FIG. First, the host command A is transmitted to the command control unit 6.
Is supplied to the host device, the host command interpreting means 13 determines that the request is an access request for the disk devices 9a, 9b, 9e, generates disk commands for the disk devices 9a, 9b, 9e, and starts / ends the disk command. It is sent to the means 15. The disk command start / end means 15 starts these disk commands.

At the same time, the host command interpreting means 1
3 judges that the disk devices 9c and 9d are unused, and notifies the prefetch control means 19 of this.
Upon receiving this notification, the prefetch control unit 19 generates a disk command to read data in the logical address area following the access area of the host command A of the disk devices 9c and 9d, and sends it to the disk command activation / end unit 15. I do. The prefetch control means 1
The fact that a read request has been made to these disk devices 9c and 9d is stored inside 9. Therefore, the disk command start / end unit 15 issues a read disk command to the disk devices 9c and 9d.

The host command A is transmitted from the disk devices a, b, and e to the old data A1 ', A2',
The new parity data P is generated by reading P ′ and performing an exclusive OR operation with the new data to generate new data A1,
The data is written to the disk devices 9a, 9b and 9e together with A2, and this is reported to the host computer 1. This completes all processing for the host command A.

At the same time, the prefetch control means 19
, The disk device 9c generated from the host command B,
9d is executed, and the old data B1 'and B2' are transferred from them, but the prefetch control means 19 makes the prefetch buffers 20c and 2c.
0d is set in a data input enabled state, and the read old data B1 'and B2' are stored here. At this time, the old data B1 'and B2' are not transferred to the data transfer means 5 'and the data buffer 7. When the transfer of the old data to the prefetch buffers 20c and 20d is completed, the prefetch control means 19 records this state in the internal storage area. The prefetch control means 1
9 is write data A1, A to the disk devices 9a, 9b.
2 and P are also controlled so that they are simultaneously stored in the prefetch buffers 20a, 20b, and 20e when they are transferred from the data buffer 7 to these disk devices 9a, 9b, and 9e.

The operation of reading the data area following the current access area and storing it in the prefetch buffer,
Saving the write data to the disk device in the prefetch buffer is collectively referred to as data prefetch.

When the host computer 1 receives the end report of the host command A, it issues the next host command B to the disk array device 2. In the command control means 6, when this is received, the host command interpreting means 13 interprets the host command B and discs 9c, 9d, 9
The disk command e is generated and sent to the disk command start / end unit 15.

The host command interpreting means 13 recognizes that the access is to the disk devices 9c, 9d, 9e.
Further, the disk devices 9a and 9b determine that they are currently unused and notify them to the prefetch control means 19. The prefetch control means 19 stores the old data B1 ', B2',
It is determined whether or not P ′ is stored in the prefetch buffers 20c, 20d, and 20e (this is referred to as a prefetch hit determination). In this case, since the old data has already been prefetched in the prefetch buffers 20c, 20d, and 20e, this is notified to the disk command activation / termination means 15 and the data transfer timing control means 17 as a prefetch hit.

Therefore, the disk command starting / ending means 15 determines that reading to the disk devices 9c, 9d, 9e is unnecessary, and issues a write disk command to the disk devices 9c, 9d, 9e at appropriate timing. The data transfer timing control means 17 sends the old data B1 ', B2', P 'and the new data B1, B2 in the data buffer 7 from the prefetch buffers 20c, 20d, 20e to the parity generation means 10 (FIG. 2). To generate a new parity P.

After the generation of the new parity P is completed, the command control means 6 receives the end notification and issues a write disk command to the disk devices 9c, 9d, 9e to write the data B1, B2 and the parity P to them. . When this write operation is completed and reported to the host computer 1, the write processing ends. FIG. 14 shows the above state.

Further, since the disk devices 9a and 9b are not executed while the disk devices 9c, 9d and 9e are performing the write operation, the prefetch control means 19 stores the next continuous data in the prefetch buffers 20a and 20b.
Then, a read disk command is generated to read the disk command, and is issued to the disk command activation / termination means 15 to prepare for the next sequential access. Further, the currently written data is stored in the prefetch buffers 20c and 20c.
0d and 20e.

As described above, when the prefetch hits, the read operation for parity generation at the time of writing becomes unnecessary, and the processing speed is doubled as compared with the conventional case.

Although the write operation has been described, it goes without saying that the sequential read operation also operates in the same manner. In this case, if all the data is stored in the prefetch buffer, the data read process can be executed without accessing the disk device at all, so that a very high-speed access without a seek time and a rotation waiting time is realized.

Further, while a certain command is being executed, a prefetch read is performed by predicting that the next continuous access will come. If another access to this disk device comes during the prefetch read, the prefetch operation is performed. Disk command start / end means 15 so that the
Should be configured. At this time, the prefetched data becomes invalid, and the access requested by the host computer 1 is immediately performed (this is called a prefetch abort).

Further, in this embodiment, the prefetch buffer 20 is provided between the data transfer means 5 'and the disk I / F 8 as shown in FIG. It does not matter.

Furthermore, in this embodiment, the prefetch buffer 20 is provided on the disk array controller 3, but may be provided inside the disk device. FIG. 15 is a block diagram showing a specific configuration of the disk device in this case, where 21 is a high-order I / F, 22 is a data buffer,
Denotes data control means, 24 denotes command control means, and 25 denotes a drive device.

Although the data buffer 22 is used here, a prefetch buffer may be provided inside the disk device 9 separately from the data buffer 22.

As described above, the prefetch control means 19
A read command for prefetch is generated by (FIG. 10), and the disk command start / end unit 15 issues a read command to the disk device 9 via the disk I / F 8. The read command is received by the upper I / F 21 and interpreted by the command control means 24, and the data is read from the drive device 25 by the data control means 23 and stored in the data buffer 22. Then, the data control means 23 reads this data from the data buffer 22 and outputs the data to the upper I / F 21 and the disk I / F 8.
The data is transferred to the disk array controller 3 (FIG. 2) via the data storage unit 3, and the data controller 5 of the disk array controller 3 may read and discard this data.

By this operation, the data is prefetched into the data buffer 22 inside the disk device 9, and when the data is accessed next time, the prefetch hits, and the disk operation is performed without a seek operation and without waiting for rotation. Data can be read from the device 9.

In this case, since the disk device 9 is accessed once, there is an overhead of the disk I / F 8, and therefore, the speed is slightly lower than that in the system in which the prefetch buffer is provided inside the disk array control device 3 described above. However, since there is no seek or rotation waiting, the speed is sufficiently higher than that of the conventional method, and the prefetch buffer is not required as compared with the above-described method in which the prefetch buffer is configured inside the disk array control device 3. At low cost. Furthermore, there is no need to perform complicated prefetch processing inside the disk array control device 3, and control is simplified. In FIG. 15, a prefetch command may be newly set instead of the method of reading and discarding the read data. When the disk command start / end unit 15 sends the prefetch command to the disk device 9, the command control unit 24 determines this and transfers the necessary data to the data buffer 22 by the data control unit 23, and ends the processing. By setting such a command, unnecessary data transfer is not performed, so that traffic between the disk I / F 8 and the upper I / F 21 can be reduced. Therefore, when a large number of disk devices are connected to the same disk I / F 8, the throughput can be increased.

As described above, according to the above two methods in which the prefetch buffer is provided in the disk device 9, the following effects can be obtained although the above effects can be obtained. That is, in a configuration in which a large number of disk devices are connected to one disk I / F 8, each disk device can perform prefetch, and thus, for example, a plurality of upper-level computers 1 issue a plurality of sequential access requests. Then, when these are operated in multiplex, if these are accesses to different disk devices, the prefetch can be hit for any of the sequential access requests from the host computer 1.

The effects of the above embodiment can be summarized as follows.

(1) When the prefetch buffer 20 is provided in the disk array device 3, if a prefetch hits at the time of writing, the operation of reading old data for parity generation becomes unnecessary, and the processing speed is reduced as compared with the conventional method. Speed up about twice.

(2) In the case where the prefetch buffer 20 is provided in the disk array device 3, if a prefetch hits at the time of reading, access to the disk device 9 becomes unnecessary, and reading is performed at the transfer speed from the prefetch buffer 20 to the host computer 1. The operation can be realized, and a very high-speed processing can be realized compared with the conventional method.

(3) If a prefetch buffer (data buffer 22) is provided in the disk device 9 and a false read command is issued from the disk array controller 3, data is prefetched into this prefetch buffer. 3, a prefetch buffer is not required, the cost is low, the control of the disk array control device 3 is easy, and the prefetch operation having substantially the same performance as the above (1) and (2) can be realized.

(4) By providing a prefetch command as a disk command, data can be prefetched into a buffer in the disk device, so that there is no need to transfer data between the disk device and the disk array control device. The traffic of the disk I / F can be reduced.

(5) A prefetch buffer (data buffer 22) is provided in the disk device 9 to perform prefetch control. In the cases (3) and (4), a plurality of sequential data access commands are issued from a plurality of upper level computers 1 substantially simultaneously. Is issued and if these are multiplexed, if these are accesses to different disk units,
For both access requests from the host computer 1,
Prefetch can be hit.

The components of the embodiment described above are:
It may be configured by either hardware or software.

The two embodiments of the command clustering control method and the sequential data prefetch control method have been described above. However, these two control methods can be performed simultaneously. At this time, the effects of both can be expected, and the effect is large in increasing the speed of data transfer.

In the above description, the parity information is stored in a single disk device (parity disk). However, the parity information is distributed to a plurality of disk devices or all disk devices. The same data control method as described above can be adopted in the storage type disk array device, and the same effect can be obtained.

Further, the clustering control method of the command has been described by taking the disk array controller as an example. However, the clustering control method is adopted in all recording / reproducing devices such as a magnetic disk controller, a magnetic tape controller, an optical disk controller, and a semiconductor disk controller. can do. At this time, an arbitrary data length can be handled, and the same effect as described above can be obtained regardless of read or write.

Further, the prefetch control method described above can be applied to all recording / reproducing apparatuses having a large number of recording / reproducing media, such as an optical disk library apparatus and a magnetic tape library apparatus, and the same effects as described above can be obtained. .

[0107]

As described above, according to the present invention,
At the time of sequential read in which n commands are continuously issued, no rotation wait occurs between commands, so that a maximum (n-1) rotation wait time can be reduced as compared with the conventional method, and the read time can be reduced. It is very effective for speeding up data transfer. Also, at the time of sequential write in which n commands for handling small data are continuously issued,
Since there is no rotation wait between commands, the maximum (2n−2) rotation waits can be reduced compared to the conventional method, and this is highly effective in increasing the speed of write data transfer.

Also, according to the present invention, during sequential write in which n commands for handling large data are successively issued, there is no rotation wait between commands, and parity data is generated. The number of read operations can be greatly reduced, so that a maximum of (2n-5) rotation waits can be reduced as compared with the conventional method, which is highly effective in increasing the speed of write data transfer.

Furthermore, according to the present invention, by providing a prefetch buffer in the disk array device, if a prefetch hits at the time of writing, a read operation for parity generation becomes unnecessary, and the processing speed is reduced as compared with the conventional method. There is an effect that the speed is approximately doubled. Also,
When a prefetch buffer is provided in the disk array device, if a prefetch hits during reading, access to the disk device becomes unnecessary, and the read operation can be realized at the transfer speed from the prefetch buffer to the host computer. There is an effect that high-speed processing can be realized.

Further, according to the present invention, when a prefetch buffer is provided in a disk device (or when realized by a data buffer), a prefetch buffer is not required in a disk array control device, and the cost is reduced. In addition, the control of the disk array control device can easily realize the prefetch operation, and the performance is the same. In this case, further, a plurality of sequential data access commands are issued from the plurality of higher-level computers at substantially the same time, and when these commands are multiplexed, if these are accesses to different disk devices, which of the higher-level computers The prefetch can be hit even for the access request of (1), and the above effect can be expected. Therefore, high throughput processing can be realized.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a first embodiment of a disk array control system according to the present invention.

FIG. 2 is a block diagram showing an example of a disk array device using a disk array control method according to the present invention.

FIG. 3 shows a logical address map of the disk array device as viewed from a host.

FIG. 4 is a diagram showing a logical address map of the disk array device after the command clustering in the first embodiment shown in FIG. 1;

FIG. 5 is a diagram showing a logical address map of each disk device in the first embodiment shown in FIG.

FIG. 6 is a timing chart showing a multiplex read operation according to a conventional method.

FIG. 7 is a timing chart showing a multiplex read operation according to the first embodiment shown in FIG. 1;

FIG. 8 is a timing chart showing another example of the multiplex write operation according to the conventional method.

FIG. 9 is a timing chart showing another example of the multiplex read operation according to the first embodiment shown in FIG. 1;

FIG. 10 shows a second example of the disk array control method according to the present invention.
FIG. 3 is a block diagram showing an embodiment.

FIG. 11 is a block diagram showing an arrangement of prefetch buffers in a disk array device using the second embodiment shown in FIG.

FIG. 12 is a diagram showing a logical address map of each disk device according to the second embodiment shown in FIG.

FIG. 13 is a timing chart showing a write operation according to a conventional method.

FIG. 14 is a timing chart showing a write operation according to the second embodiment shown in FIG.

FIG. 15 is a block diagram showing an example of the internal structure of the disk device according to the second embodiment shown in FIG.

FIG. 16 is a block diagram showing an example of a conventional disk array control method.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Upper computer 2 Disk array device 3 Disk array controller 4 Upper I / F 5 Data control means 6 Command control means 7 Data buffer 8a-8e Disk I / F 9a-9e Disk device 10 Parity generation means 11 Host command input / output means 12 Host Command Queuing Means 13 Host Command Interpreting Means 14 Command Clustering Means 15 Disk Command Start / End Means 16 Main Control Means 17 Data Transfer Timing Control Means 18 Disk Command Management Table 19 Prefetch Control Means 20a-20e Prefetch Buffer

──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Takashi Oeda 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture Inside Hitachi, Ltd. Microelectronics Equipment Development Laboratory (72) Inventor Seishi Honda Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa 292 Hitachi Electronics, Ltd. Microelectronics Equipment Development Laboratory (58) Field surveyed (Int.Cl. ⁶ , DB name) G06F 3/06

Claims (12)

(57) [Claims] A means for receiving a command from a host computer and notifying the end of the command; a means for interpreting a command issued from the host computer and converting the command into a command for each disk device; Means for initiating a command to the device and detecting completion of read and write data transfer by the command; means for controlling data transfer between the host computer and each of the disk devices; A control method of a disk array control device comprising: means for controlling, a command for temporarily storing commands continuously issued by the host computer in a queue and managing the commands; and a plurality of commands stored in the queue. When the command requests access to one continuous area, these commands are grouped into one command, and Means for converting one looped command to a command to each disk device; conversion information for converting one grouped command to a command to each disk device; Means for storing and managing information data for the commands, wherein a plurality of commands to the continuous area issued by the upper-level computer are regarded as one sequential access command in the disk array controller, so that each Data handled by the command is treated as one continuous data,
A disk array control method wherein the entire data is controlled so as to distribute and issue a command to each of a plurality of disk devices for access. 2. The disk array control device according to claim 1, wherein a plurality of read commands to the continuous area issued by the host computer are regarded as one read command inside the disk array control device, and the plurality of read commands are determined based on the one read command. A disk array control method, wherein a read command is distributed and issued to each of the disk devices so that all data handled by the plurality of read commands to the continuous area are controlled to be read collectively. 3. The disk array control device according to claim 1, wherein a plurality of write commands to the continuous area issued by the host computer are regarded as one write command inside the disk array control device, and based on the one write command, By distributing and issuing a read command to each of the plurality of disk devices, an area in which all data handled by the plurality of write commands is to be written to the continuous area is read from the data and parity information already recorded therein. And collectively reads the read data, generates new parity information from the read data, the parity information and the data to be written, and then writes a write command to each of the disk devices based on the one write command. Distribute, issue, and write all of the data to be written and the new parity information generated. The disk array control method and controls to write and Batch. 4. The disk array control device according to claim 1, wherein a plurality of write commands to the continuous area issued by the host computer are regarded as one write command in the disk array controller, and the data of the write command is to be written. A data area in which data is to be written to all stripes where writing is started from the same logical address in all the disk devices is cut out, and the data area is written to the data area without reading old data from the data area. By generating parity data from the data and writing the data and the parity data collectively, the data area is controlled so as to eliminate reading of old data for generating the parity data. Characterized disk array control method. 5. The disk array according to claim 1, further comprising: first control means for controlling so that continuous data following data handled by a command issued by the host computer is pre-read from the disk device. Data storage means for storing data prefetched by the first control means in a control device or in each of the disk devices; and second control for controlling input / output of the prefetched data to / from the data storage means. Means for controlling a disk array. 6. The storage device according to claim 5, further comprising: a determination unit configured to determine whether data handled by the read command issued from the host computer is stored in the data storage unit. When it is determined that the data handled by the read command is stored in the means,
A disk array control method wherein the data is read from the data storage means and sent to the host computer. 7. The disk array control method according to claim 5, wherein parity information generated for a write command issued from the host computer is stored in the data storage unit. 8. The data storage means according to claim 5, wherein old data and old parity information already recorded in an area to which new data handled by a write command issued from a host computer is to be written are stored in said data storage means. Determining means for determining whether the old data and the old parity information are stored in the data storage means, extracting the old data and the old parity information from the data storage means A new parity information is generated from the new data, the old read data and the old parity information, and the new data and the generated new parity information are controlled to be written to a disk device. Array control method. 9. The read command according to claim 5, 6 or 8, wherein a read command is issued to a disk device that has not been an access target for storing data in an area subsequent to the data area requested by the host computer, A disk array control method wherein the read data is stored in the data storage means, and data in an area following a data area of an access request from the host computer is read in advance. 10. The data reading device according to claim 9, wherein the data read from the disk device is discarded, whereby the data is transferred in advance to a data buffer in the disk device, and the next time a read access request is made to the area. A data read from said data buffer. 11. The disk device according to claim 5, 6 or 8, wherein a disk device that has not been an access target for storing data in an area subsequent to the data area requested by the higher-level computer is provided inside the disk apparatus. A disk array control method, comprising: a data prefetch command for controlling data to be read to a certain data buffer; and prefetching data in an area following a data area of an access request from the host computer. 12. The disk drive according to claim 9, 10 or 11, wherein when there is an access request to the disk device from the host computer during read command processing for read-ahead to the disk device. A disk array control method for performing control so as to stop read processing.