JP2000242441A - Disk control method and its controller - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent the performance of access to an area of high priority from decreasing owing to access to other areas by setting as reference values by the areas of a disk one of an access processing frequency, an access data amount, and an access time and accessing the respective areas according to the reference values. SOLUTION: In addition to a setting means which sets reference values for the access frequencies, access data amounts, etc., of the areas A and B of the disk according to access priority, a disk control part 4 is provided with a disk access control means 4a which adjusts the access frequencies and access amounts by the areas A and B according to the set values of the areas A and B set by the setting means. The two areas A and B of the disk 3 are given priority. Namely, data of high priority are recorded in the area A and data of relatively low priority are recorded in the area B. According to the priority, a reference value for an access ratio is set in advance and a queue processing by access is performed according to the reference value.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a disk control method and a disk control device for improving the access performance of a disk area.

[0002]

2. Description of the Related Art A disk control device includes an OS (Operating System) for controlling the operation of the entire device, a main control unit for outputting an access request in accordance with an external instruction, a still image
The disk includes a disk for recording image data such as a moving image, a disk controller (disk driver / disk controller) for sequentially writing and reading data to and from a disk area in response to an access request from the main controller.

In the random access from the main control unit to the disk area, the data read / write head is frequently moved (seeked) every time the access occurs. As a result, there is a problem that the performance of data read / write speed is reduced.

[0004] Therefore, the disk control unit is arranged to execute an access request rearrangement process after accepting an access request from the main control unit, so as to minimize seeks.

[0005]

However, it is difficult for such a disk control device to appropriately control the data read / write speed performance.

Hereinafter, the reason will be described. Now, for example, as shown in FIG. 7, two data areas A,
Assume that B exists. The area A is arranged in a small area at the head of the disk, and stores frequently accessed data. The area B has a larger area than the area A, and stores data with a low access frequency.

In such a case, in the case of accessing only the area A, the occurrence of seek is small and high performance can be obtained. However, in the case of accessing the area B, more seeks are performed than in the case of accessing the area A. Occurs, affecting the performance.

Therefore, in order to maintain high performance, a plurality of disks are prepared, and the area A is provided at the head of one disk.
It is considered that the area B is left empty by disposing the area B, the area B is arranged at the head of the other disk, and data corresponding to the area B shown in FIG. 7 is read and written. But,
This idea requires a trade-off between cost and performance, and is not an appropriate improvement method.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a disk control method and a disk control device for preventing access to another area from deteriorating access performance in a high priority area. .

[0010]

In order to solve the above problems, the present invention sets, for each area of a disk, one of an access processing count, an access data amount, and an access time as a reference value, This is a disk control method for executing access processing to each area based on this reference value.

According to such a control method, for example, the number of times of access processing to each area of the disk is performed based on the ratio of a predetermined number of times of access. The access performance in the high area does not deteriorate.

Another aspect of the present invention is a disk in which an area for storing data having a high access priority and an area for storing data having a low access priority are set, and an access process is performed in accordance with the access priority of each area of the disk. A disk control comprising: setting means for setting at least one of the number of times, access data amount, and access time as a reference value; and disk access control means for monitoring an access request and executing access to the disk based on the reference value. Device.

According to the present invention, by taking the above measures, data having a high access priority is stored in an area at the head of the disk, and data having a low access priority is stored in an area lower than the area at the head. In addition, if the ratio of the number of times of access processing is set according to the priority, the access processing of each area can be executed according to the ratio of each area of the disk, and the access performance of data with high access priority is reduced. Disappears.

Further, another invention measures the number of access processes per unit time and the like for each area of a disk,
This measurement value can be automatically set as a reference value such as the number of times of access processing, and the above means can be easily applied to a RAID system.

[0015]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of a disk control device according to the present invention.

The disk control device includes a memory 1, a main control unit 2, a disk 3, and a disk control unit 4 for controlling the disk 3 and controlling data access between the main control unit 2 and the disk 3. .

The memory 1 comprises an OS (Operating System) for controlling the basic operation of the entire apparatus, various necessary programs, and a ROM and / or a RAM for storing control information necessary for executing each program. The RAM is used exclusively as a data work area for realizing a predetermined process in the main control unit 2.

The main control unit 2 outputs an instruction signal input from an input device 5 such as a keyboard or a pointing device or an access request required in the course of a predetermined program processing, and converts the original data stored in the memory 1 Disk 3
Write data to a certain area or read data from a certain area of the disk 3 and store it in the memory 1 to perform processing such as data restoration.

The disk control unit 4 has a function of controlling reading and writing of information from and to the disk 3 loaded in the disk drive when receiving an access request from the main control unit 2.

The disk controller 4 is set in advance by setting means for setting reference values such as the number of accesses to each area A and B of the disk 3 and the amount of access data in accordance with the access priority. A disk access control means 4a is provided for adjusting the number of accesses and the amount of access for each area of the disk 3 according to the set values of the areas A and B.

The disk access control means 4a includes, for example, the respective areas A and B of the disk 3 as shown in FIG.
A counter 11 that counts the number of times access is performed
A, an area B counter 11B and queues 12A, 12B for each area A, B are provided.
It has a function of executing a process as shown in FIG. 3 based on the disk access control program stored in 1.

In addition to the image input device 6, the disk control device includes a display device 8 for displaying input data, data of a processing process by the main control unit 2, data of a processing result, and the like via a display control unit 7. Is provided. The image input device 6 includes a scanner for reading a document, an image, and the like, a camera for capturing an image of a subject and outputting image data representing a still image or a moving image, a recording medium for recording image data in advance, a predetermined medium, A communication control device that receives image data from another device in the transmission control procedure described above is used.

Next, a description will be given of a processing procedure of the disk control unit 4 which is an essential part of the present invention, among the devices configured as described above.

In the conventional disk control unit, when a data write request and a data read request are received from the main control unit for each area of the disk, those access requests are sequentially arranged in one queue, and the processing is performed in the arrangement order. The queues are rearranged so that address areas close to each other are gathered in order to reduce seeks, and processing is performed according to the rearrangement.

On the other hand, as the control device of the present invention,
When the disk 3 is divided into, for example, two areas A and B, the user assigns priority to the two areas A and B of the disk 3 and records high-priority data in the head area A,
Data having a relatively low priority is recorded in an area B lower than the head area, a reference value of an access ratio is set in advance according to the priority of each area, and a queue process by access is executed according to the reference value. Is to do.
For example, the ratio of the number of access processes of the queues 12A and 12B corresponding to the areas A and B is set to 9: 1. In this case, the area queues 12A and 12B are selected and processed so that the access processing of the area B is performed once while the access processing of the area A is performed nine times.

Hereinafter, a specific processing procedure by the disk control unit 4 will be described with reference to FIG.

The disk controller 4 performs an initialization process.
In this initialization process, unnecessary data is cleared, and the user sets the number of times of queue processing for the areas A and B, that is, the ratio of the number of times of access processing (S1). In this case, the area A
, The access processing count is set to be higher than that of the area B.

After the number of access processes (reference value) for each of the areas A and B is set as described above, when an access request from the main control unit 2 to the disk device 3 is made, the request target areas are A and B. (S2, S3), and according to the confirmation result, the queue 12A corresponding to the corresponding area,
12B queues the access request contents (S4,
S5).

Thereafter, the disk control unit 4 checks the count value of the area A counter 11A (S6). If the count value is "0", it is determined whether there is access request data in the area A side queue 12A. Is checked (S7). Queue 1
If there is access request data in 2A, the queue 12A
Then, the access request data is read from the server, and the access request is processed (S8, S9). Then, the area A side queue 12A
After the access processing, the count value of the A counter 11A is increased by +
After incrementing by 1 (S10), it is determined whether or not the A-side set value (for example, 9) has been reached (S11). If not, the process returns to step S7 again, as long as the access request data remains in the queue 12A. , Access processing is performed until the A-side set value is reached.

In step S6, the count value of the area A counter 11A is "0", but there is no data in the A-side queue 12A, or the count value of the A counter 11A is other than "0", and When there is no data in the queue 12A, and when the value exceeds the A-side set value in step S11, the process proceeds to step S12, and it is determined whether the count value of the area B counter 11B is "0".
Here, when the count value is “0”, the queue 12B
Then, the access request data is read from and the access request is processed (S13, S14). Then, the area B side queue 1
After the 2B access processing, the count value of the B counter 11B is incremented by +1 (S15), and then it is determined whether or not the area A counter 11A exceeds the set value on the A side (S15).
16) If not, the process proceeds to step S7, and the process is repeated as described above.

In step S16, the area A counter 1
If 1A exceeds the A-side set value, the process proceeds to step S17, and it is determined whether or not the count value of the area B counter 11B has reached the B-side set value. If the B-side set value is "1", the count value of the area B counter 11B is reached. In this case, the areas A and B counters 11A and 11B are cleared (S18), and step S6 is performed.
And the same processing is repeated.

Incidentally, the case where the B-side set value is "2" or more is naturally expected. In this case, it is determined in step S17 that the B-side set value has not been reached, and the flow shifts to step S13 to repeat the same processing.

If the count value of the area B counter 11B is other than "0" in step S12, it is determined whether or not the count value has reached the B-side set value (step S19). Step S in case
If it has reached 13, the process proceeds to step S16.

Therefore, the disk control unit 4 controls each area A,
Since the access processing is performed in accordance with the access processing frequency ratio of each area set in advance based on the priority of B, the access processing of the area B does not reduce the access processing of the area A having a high priority. The occurrence of seek can be minimized.

Further, another disk access control means 4a
As an example, the access processing of each disk area A, B may be performed based on the number of access processing times per unit time. In this case, for example, the number of access processes per unit time of the low priority area B is set to, for example, 10 per second.
A method such as limiting the number of times can be considered.

Specifically, from a certain time T, T + t (t <
When the access processing of the area B is executed 10 times during (1 second), the access processing of the area B is stopped until the time of T + 1 seconds, and the access processing of the area A is exclusively executed during the remaining time.

Therefore, according to this disk control device,
Since the number of access processes per unit time of the low-priority disk area is limited, the high-priority disk area A
Access performance is not degraded.

Further, a time ratio between the access processing time of the area A and the access processing time of the area B may be set, and the access processing of each area may be executed according to the time ratio of each area.

Although the above-described disk control device has described an example in which the access processing is controlled by the number of times of access processing or the number of times of access processing or the ratio of time, for example, the ratio of the amount of access data is determined according to the priority of the areas A and B. The access processing may be controlled using a similar processing procedure.

One is that, as shown in FIG. 4, data is stored in each of the areas A and B of the disk 3 by dividing the data into blocks. This is an example of adjusting the access processing of each area according to the ratio of the amount of access data.

For example, in a certain unit time, 9
While performing the access processing on the 0 block, the access processing is performed on the 10 blocks in the area B.
The access processing is executed by selecting the queues 12A and 12B of each area. In this case, the ratio of the amount of access data per unit time is 90 blocks: 10 blocks,
It can be realized by a processing procedure similar to that of FIG.

As an example of limiting the amount of access data processing per unit time, for example, it is conceivable to limit the amount of access data in a low priority area.

More specifically, with respect to the area B, 1 per second
For example, it is limited by 00 KBytes. That is, if the amount of access data to the area B exceeds 100 KBytes between T + t (t <1 second) from a certain time T, T + 1
The access processing of the area B is stopped until the time of seconds is reached, and the access processing of the area A is exclusively executed during the remaining time.

Therefore, according to this disk control device,
Since the amount of access data per unit time of the low-priority disk area is limited, the high-priority disk area A
Access performance does not decrease.

Further, when it is necessary to give priority to the access processing of the queue 12A corresponding to the area A, the access processing of the queue 12A corresponding to the area B having the lower priority is maintained so that the ratio of the access processing does not collapse. Does not split even if the amount of access data is large,
In the case of the queue 12B, when one access data amount is large, it is conceivable to perform processing by dividing the data.

(Other Embodiments) (1) In the above embodiment, the user sets the number of times of access processing, the amount of access data, and the time according to the priority of each area of the disk 3. It is also possible to automatically set the number of accesses and the like.

In the disk control unit 4, for each of the areas A and B of the disk 3, a measuring means for measuring any one of the number of accesses per unit time this time, the amount of access data and the access time, and this measuring means In addition to the automatic setting means for automatically setting the number of accesses within the next unit time, the amount of access data, and the access time using the measurement result measured in the unit time this time, the number of accesses to each input disk area, Disk access control means 4a for monitoring the data amount or access time and executing access processing for each area of the disk according to the number of accesses to each area, the access data amount, the access time or the ratio set by the setting means is provided. Have been.

Hereinafter, one specific example will be described. An A frequency counter and a B frequency counter are provided in the memory 1 or in the disk control unit 4, and the number of access processes to the areas A and B is counted within a certain unit time. Then, when a certain unit time is reached, the ratio of the number of access processing times is calculated using the count value of the dual frequency counter, and set as the ratio of the number of queue processing times of the areas A and B, that is, the ratio of the number of access processing times. During the next unit time, the access processing of the area A and the area B is executed based on the set ratio of the number of access processing, while the contents of the A frequency counter and the B frequency counter are cleared, and the next unit time is cleared. In order to calculate the ratio of the number of times of access processing, the number of times of access processing to areas A and B is newly counted.

As described above, the ratio of the number of access processing times in the area A and the area B is automatically set while monitoring the access frequency within the previous unit time for each unit time. The degree of priority is set low, and in the case of an area with a high access frequency, the priority is set high, so that access processing can always be executed with the optimum priority.

In the above example, the number of times of access processing of each area per unit time is measured. For example, the amount of access data or the access time is measured, and based on the measurement result, the next unit time within the next unit time is measured. A method of setting an access data amount or access time of each area or a ratio thereof may be used.

(2) In the above embodiment, the access processing of each of the areas A and B of the single disk 3 has been described. However, a RAID system that executes access processing by regarding a plurality of disks as virtually one disk Is similarly applicable.

This RAID is classified from level 0 to level 6. Here, the application of RAID 0 which becomes level 0 will be described. Note that RAID 0 is a technique for distributing load, also called striping, in which a plurality of disks are connected to increase the capacity and virtually configure a large-capacity disk. This RAID 0 includes, for example, as shown in FIG. 5, the physical areas a1, a2, and 3 of _three real physical disks 3 ₁ , 3 ₂ , and 3 ₃ .
The capacity is increased by treating a3 as one virtual area.

The disk controller 4 is configured, for example, as shown in FIG. That is, for example, as shown in FIG. 3, an application 21 for realizing the disk access control means 4a for adjusting the number of times of access processing for each area and the like, and when an application request for access to a certain area is received from this application 21, Table 22
Driver 23 which specifies one disk request area among a plurality of disks while watching and outputs drive control data for access processing.
And a disk driver 24 for executing access processing to a requested area of a certain disk in accordance with drive control data from the disk drive.

[0055] The address conversion table 22 in advance of the region A of the virtual disk address A1 actual physical disk 3 ₁ address, the address A1 and the actual physical disk 3
₂ and so on.

On the other hand, each disk (real physical disk)
3 _{1, 3} 2, _{3 3} side is divided into two regions, respectively, that is the actual physical area a1 b1, a2 and b2, a3 and the b3.

In such a case, the virtual disk is
Real physical disks a1, a2 and a configured by D0
3 and a virtual disk area B of real physical disks b1, b2, and b3 also configured by RAID0.

In the disk control apparatus applied to such a RAID system, the virtual driver 23 executes the access processing to the areas A and B executed in the process shown in FIG. 3 based on the RAID software. Since the access processing is performed by applying the above, the same functions as those of the above embodiment can be provided, and the same effects can be obtained.

Therefore, according to such an embodiment,
Using a plurality of small-capacity disks, it is possible to appropriately execute access processing for each area of each disk while improving the apparent performance of the disk.

In this embodiment, RAID0 has been described, but it goes without saying that the same can be applied to other RAID1 to RAID6.

[0061]

As described above, according to the present invention, there is provided a disk control method and a disk control apparatus capable of executing access processing without deteriorating access performance of a high priority area by accessing another area of the disk. it can.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing an embodiment of a disk control device according to the present invention.

FIG. 2 is an exemplary view for explaining hardware components of the disk control unit shown in FIG. 1;

FIG. 3 is a flowchart illustrating the operation of the disk access control device according to the present invention.

FIG. 4 is a view for explaining data blocks in each area in the disk.

FIG. 5 is a view for explaining an example in which data areas of a plurality of disks are treated as one area.

FIG. 6 is a diagram showing a configuration example of a disk control unit when data areas of a plurality of disks are treated as one area.

FIG. 7 is an area layout diagram of a disk for explaining conventional disk control.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 2 ... Main control part 3 ... Disk 2 ... Parameter generation mechanism 4 ... Disk control part 4a ... Disk access control means 21 ... Disk access control application 23 ... Virtual driver 24 ... Disk driver

Claims (4)

[Claims] 1. A disk control method for performing an access process for each area of a disk, wherein one of an access processing count, an access data amount, and an access time is set as a reference value for each area of the disk. A disk control method, wherein an access process to each area is executed based on a reference value. 2. A disk in which an area for storing data with a high access priority and an area for storing data with a low access priority are set, and the number of access processes and the amount of access data according to the access priority of each area of the disk And a setting means for setting at least one of the access times as a reference value, and disk access control means for monitoring an access request and executing access to the disk based on the reference value. Control device. 3. A measuring means for measuring any one of the number of times of access processing per unit time, an access data amount, and an access time for each area of a disk, and a measurement which is measured by the measuring means in a unit time this time. Automatic setting means for automatically setting at least one of the number of times of access processing in the next unit time, the amount of access data, and the access time as a reference value using the result; monitoring an access request; A disk control device, comprising: disk access control means for performing access to a disk. 4. The disk controller according to claim 2, wherein each area of the plurality of disks is a virtual disk area configured by a required level of RAID,
When receiving an instruction to access each area from the disk access control means, the virtual disk area identifies one requested disk area among a plurality of disks and executes an access processing to the specific disk area. A disk control device provided with a driver.