JP2002140169A - Disc array control unit and disc array control method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a disc array control unit and disc array control method capable of enhancing the response time of input and output to a disc having a performance difference to improve the response time of the whole RAID. SOLUTION: This control unit comprises a means for assigning, according to the response time of each device connected to a disc array device, part of a cache capacity on a disc array exclusively to the device.

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 technique, and more particularly, to a disk array control device and a disk array for improving the response time of an entire RAID by accelerating the input / output response time for disks having performance differences. It relates to a control method.

[0002]

2. Description of the Related Art As an array type storage system, a disk array system is generally used. A disk array system is a computer system that achieves high data transfer speed and high reliability between a processing device and an external storage device by operating a plurality of magnetic disk devices in parallel in a computer system. .

As an example of the configuration of a disk array system, a paper by Paterson et al.
n, G. Gibson, R .; Katz, "A Case
for Redundant Arrays of
Inexpensive Disks (RAID) ",
ACM SIGMOD conference pro
ceedings, 1988, pp. 109-11
6). In the above paper, RAID (Redu
ndant Array of Inexpensive
e Disks) (Raid).

In the RAID, ordinary input / output data is distributed and stored in a plurality of drives, and redundant data is stored. This redundant data installation can be used together with a data restoration function when data is lost,
The purpose is to allow the missing data to be restored.

[0005] Cache memories in disk controllers have been introduced to improve the performance gap between processing and storage devices. With the introduction of the cache memory, the processing time of the write request from the processing device is reduced from the time required for recording data to the medium of the storage device to the time required for writing the data to the cache memory, which has been greatly reduced. Further, the processing time of the read request can be reduced by hit processing or pre-read processing in the cache memory.

[0006] The use form of the cache memory is broadly divided into integrated use and divided use.

In the form of integrated use, one cache memory is arranged in a control device, and this cache memory handles all data under the control device. This format is
A large-capacity memory can be used collectively, which is effective from the viewpoint of memory utilization.

On the other hand, in the form of divisional use, a plurality of cache memories are arranged in a control device, and each cache memory is in charge of a separate storage device. This format has the advantage that it is easily applied to individual data access patterns. On the other hand, since the cache is subdivided and used, wasteful memory is generated, and the memory utilization efficiency is poor.
Further, there is another problem that it is necessary to adjust the data arrangement to the cache arrangement in order to cope with the difference in the access pattern.

[0009]

However, in the above prior art, RAID (Redundant Ar) is used.
rays of Inexpensive Disk
Each of the disks constituting s) has some individual difference even if it is designed to have the same rotation speed.
This effect appears as a response time at the time of Read / Write, and increases as the load increases, and a difference of several milliseconds may occur in the average response time in some cases.

In particular, in RAID0 and the like, the performance of the slowest disk among the constituent disks greatly affects the determination of the performance of the RAID as a whole. Even when disks with the same rotation speed are used, there is a problem that overall performance may be deteriorated when devices of poor quality are included.

The present invention has been made in view of the above problems, and has as its object to speed up the response time of input / output to / from a disk having a difference in performance and improve the response time of the entire RAID. It is to provide a disk array control device and a disk array control method.

[0012]

According to a first aspect of the present invention, a part of a cache capacity on a disk array is assigned to a device according to a response time of each device connected to the disk array device. A disk array control device characterized by having means for exclusive assignment. The gist of the invention described in claim 2 of the present invention is as follows.
Means for determining the performance difference between devices by calculating the average value of the response times of the devices connected to the disk array within a certain period of time during actual operation and comparing the response times 2. The disk array control device according to claim 1, wherein The gist of the invention described in claim 3 of the present invention is that, for a disk having a predetermined performance difference, means for allocating an arbitrary cache capacity exclusively to the disk based on the individual performance difference is provided. 3. The disk array control device according to claim 2, wherein Further, the gist of the invention described in claim 4 of the present invention is that, for a disk having a predetermined performance degradation, means for allocating an arbitrary cache capacity exclusively to the disk based on the individual performance difference is provided. 3. The disk array controller according to claim 2, wherein: The gist of the invention described in claim 5 of the present invention is characterized in that there is provided means for recording an error occurrence rate of each drive and selecting a drive based on the recorded error occurrence rate. Or the disk array control device described in 4. The gist of the invention described in claim 6 of the present invention resides in the disk array control device according to any one of claims 3 to 5, further comprising means for allocating a dedicated area for a cache. The gist of the invention described in claim 7 of the present invention is as follows.
The disk array control device according to any one of claims 3 to 6, further comprising means for controlling a disk cache based on the number of reduced logical volumes included in the disk array. The gist of the invention described in claim 8 of the present invention consists of a control unit, a response time measuring unit, a cache control unit, and a cache.
Connected through an interface between the PUs and the CP
A disk array controller that receives a Read / Write command from the U via the interface;
Read / write connected to the disk array controller via an I / O interface and an I / O bus.
A response time measuring unit including a counter, an average value calculating unit, a device table, a response time buffer, and a cache judging unit; and a response / read instruction from the control unit. Upon receiving the issuance, the counter calculates the time from the issuance of an instruction to each device to the response, inputs the response time of the instruction to the response time buffer, and stores in the device table up to the immediately preceding instruction for each device. The average value of the response time and the number of instructions are stored, and the average value calculation unit stores the {m-1} stored in the device table.
The average value of the response time up to the m-th instruction is obtained from the average value of the response time up to the m-th instruction and the response time of the m-th instruction held in the response time buffer. The cache determination unit performs a cache allocation determination for each device based on the information in the device table, registers the device to be allocated in the device table, and the cache control unit 8. The disk array control device according to claim 3, wherein the cache is controlled in accordance with cache allocation information included in the device table. The gist of the invention described in claim 9 of the present invention is to have a step of allocating a part of the cache capacity on the disk array exclusively to the device according to the response time of each device connected to the disk array device. The disk array control method is characterized in that: The gist of the invention described in claim 10 of the present invention is that, for a device connected to a disk array, an average value of response times of respective devices within a certain period of time during actual operation is obtained and the comparison is performed. 10. The method according to claim 9, further comprising the step of determining a performance difference between the devices. Further, the gist of the present invention described in claim 11 of the present invention has a step of, for a disk having a predetermined performance difference, allocating an arbitrary cache capacity exclusively to the disk based on the individual performance difference. A disk array control method according to claim 10, wherein: The gist of the invention described in claim 12 of the present invention is that, for a disk having a predetermined performance deterioration, a step of allocating an arbitrary cache capacity exclusively to the disk based on the individual performance difference. 2. The method according to claim 1, wherein
0 in the disk array control method. The gist of the invention described in claim 13 of the present invention is characterized in that it has a step of recording an error occurrence rate of each drive and selecting a drive based on the recorded error occurrence rate. Or the disk array control method described in 12. The gist of the invention described in claim 14 of the present invention resides in a disk array control method according to any one of claims 11 to 13, further comprising a step of allocating a dedicated area for a cache. Also,
The gist of the present invention described in claim 15 of the present invention is that the method further comprises a step of controlling the disk cache based on the number of reduced logical volumes included in the disk array. In the disk array control method described above.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS According to the present invention, a large cache capacity can be obtained by allocating a part of the cache capacity on a disk array exclusively to a device according to the response time of each device connected to the disk array device. The response time of the entire disk array can be improved by reducing the difference between input and output to each device. Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

(First Embodiment) Hereinafter, a first embodiment of the present invention will be described.
An embodiment will be described in detail with reference to the drawings. FIG.
FIG. 2 is a functional block diagram for explaining the disk array control device 20 according to the first embodiment of the present invention. In FIG. 1, 1 is a CPU, 2 is an interface, 3 is a controller, 4 is a counter, 5 is an average calculator, 6 is a cache controller, 7 is a device table, 8 is a response time buffer, and 9 is a disk array controller. (Indicated as DAC in the figure) 10 is a cache, 11 is a response time measuring unit, 12
Denotes an I / O interface, 13 denotes an I / O bus, 14 denotes a cache determination unit, and 20 denotes a disk array control device according to the present embodiment.

Referring to FIG. 1, in a disk array control device 20, a CPU 1 includes an I / O interface 1
2 is a higher-level device that is connected to the disk array controller 9 (DAC) via the I / O bus 13 and issues a Read / Write command to the disk array controller 9 (DAC). The disk array controller 9 (DAC) mainly includes the control unit 3, the response time measurement unit 11,
A plurality of disks (specifically, a first disk and a second disk shown in FIG.
Disk, 3rd disk, ..., Nth disk) and CPU
1 are connected via an interface 2.

The response time measuring section 11 comprises a counter 4, an average value calculating section 5, a device table 7, a response time buffer 8, and a cache determining section 14. In response to the issuance of the Read / Write command from the control unit 3, the counter 4 calculates the time from the command issuance to each device to the response.

The device table 7 holds the average value of the response time up to the previous command for each device and the number of commands.

In the average value calculation section 5, the device table 7
The response time from the response time of the m-th instruction held in the response time buffer 8 to the average value of the response time up to the {m-1} -th instruction held in The average value of the time is obtained, and the information of the device table 7 is updated.

The cache determination unit 14 determines cache allocation to each device based on the information in the device table 7 and registers the device to be allocated in the device table 7.

The cache control unit 6 controls the cache 10 in accordance with the cache assignment information included in the device table 7.

Next, the operation of the disk array controller 20 (disk array control method) will be described. FIG.
FIG. 3 is a memory map (part 1) for explaining the operation of the first embodiment, and FIG. 3 is a memory map (part 2) for explaining the operation of the first embodiment.

In this embodiment, as shown in FIG. 1, when a Read / Write command is issued from the control unit 3, the time from the command issuance to the response is counted.
The response time of the instruction is held in the response time buffer 8.

The device table 7 holds the average value of the response time up to the {m-1} th time and the number of instructions m-1. The data and the m-th instruction held in the response time buffer 8 are stored. Is input to the average value calculation unit 5, and the average value of the response time up to the m-th instruction and the number m of instructions obtained from the result are held in the device table 7 again.

The cache determination unit 14 compares the average values of the response times of all the devices held in the device table 7 at certain time intervals, and selects a device having a remarkably poor response time as a cache allocation target.
Register in the device table 7. The disk registered in the device table 7 (specifically, the first disk shown in FIG. 1)
As shown in FIG. 2, a part of the cache 10 is used as a disk cache for the disks, the second disk, the third disk,..., The Nth disk.

A device having a response time worse than the average is N
When there are (≧ 2) units, as shown in FIG. 3, the total capacity allocated as a disk cache (the upper limit of the total amount is indicated as the allocation upper limit in FIG. 3) is divided into N equal parts (see FIG. 3). The disk allocating unit 1,..., The disk allocating unit N is used for each disk (specifically, the first disk shown in FIG. 1).
Disk, second disk, third disk,..., Nth disk). The total capacity of the cache 10 allocated to the relevant disk is set to an arbitrary upper limit, and is not set to a value larger than the upper limit. In addition, each disk (specifically, the first disk, the second disk, the third disk,.
The cache capacity allocated to the disk is assumed to be larger than the capacity of the cache 10 of the disk itself, and when the capacity divided into N equal parts is smaller than that, priority is given to the one with slower response time. It is assumed that cache allocation is to be performed on a case-by-case basis, and that cache allocation beyond the upper limit is not performed.

Of the data that is accessed to the striped RAID, data of a portion determined by the cache control unit 6 referring to the device table 7 to be an access to the disk to which the cache is to be allocated is a normal data. The access is performed not through the cache 10 but through the dedicated cache 10.

As described above, according to the first embodiment, the device connected to the disk array
The average value of the response time of each device within a certain period of time during actual operation is determined, and the comparison is performed to determine a performance difference between the devices and to determine a disk having a certain level of performance difference (performance degradation). , An arbitrary cache capacity is allocated exclusively to the disk based on the individual performance difference. As described above, by using the cache 10 which is dedicated to a large amount, I / O to a slow disk is reduced, the average response time is shortened, and the actual response time of the configured RAID is improved. As a result, the response time of input / output to / from a disk having a performance difference (deterioration in performance) can be shortened, and the response time of the entire RAID can be improved.

(Second Embodiment) Hereinafter, a second embodiment of the present invention will be described.
An embodiment will be described in detail with reference to the drawings. FIG.
FIG. 7 is a functional block diagram for explaining a disk array control device 20 according to a second embodiment of the present invention.

In the first embodiment, the cache 10 is dynamically allocated using the cache determination unit 14. On the other hand, the present embodiment is characterized in that the cache 10 is fixedly assigned during operation by an external input. As shown in FIG. 4, in the present embodiment, the average response time of each device stored in the device table 7 can be externally referred to, and the value is used as a reference to allocate the cache 10 by the external input. Select the target device and set the device table 7
And register it as a target device for cache allocation.
The cache control unit 6 controls the cache 10 with reference to the registered settings. Thereby, the same effect as that of the first embodiment can be realized.

It is apparent that the present invention is not limited to the above embodiments, and that the above embodiments can be appropriately modified within the scope of the technical idea of the present invention. Further, the number, position, shape, and the like of the constituent members are not limited to the above-described embodiments, and can be set to numbers, positions, shapes, and the like suitable for carrying out the present invention. In each drawing, the same components are denoted by the same reference numerals.

[0031]

According to the present invention, the average value of the response times of the devices connected to the disk array within a certain time during the actual operation is obtained, and the average value is compared. In addition to determining the performance difference, for a disk having a certain degree of performance difference (deterioration in performance), an arbitrary cache capacity is allocated exclusively to the disk based on the individual performance difference, and the performance difference (performance (Deterioration of the disk), the response time of input / output to and from the disk where the deterioration of the disk is observed can be shortened, and the response time of the entire RAID can be improved.

[Brief description of the drawings]

FIG. 1 is a functional block diagram for explaining a disk array control device according to a first embodiment of the present invention.

FIG. 2 is a memory map (part 1) for explaining the operation of the first embodiment;

FIG. 3 is a memory map (2) for explaining the operation of the first embodiment;

FIG. 4 is a functional block diagram for explaining a disk array control device according to a second embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... CPU 2 ... Interface 3 ... Control part 4 ... Counter 5 ... Average value calculation part 6 ... Cache control part 7 ... Device table 8 ... Response time buffer 9 ... Disk array controller 10 ... Cache 11 ... Response time measurement part 12 ... I / O interface 13 I / O bus 14 Cache determination unit 20 Disk array controller

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G06F 12/08 557 G06F 12/08 557

Claims (15)

[Claims] 1. A disk array control device comprising means for allocating a part of cache capacity on a disk array exclusively to a device according to a response time of each device connected to the disk array device. 2. A device connected to a disk array is averaged in response time of each device within a fixed time during actual operation, and a comparison is made to determine a performance difference between each device. 2. The disk array control device according to claim 1, further comprising means for performing. 3. A disk according to claim 2, further comprising means for allocating an arbitrary cache capacity exclusively to the disk having a predetermined performance difference based on the individual performance difference. Disk array controller. 4. The apparatus according to claim 2, further comprising means for allocating an arbitrary cache capacity exclusively to the disk on which a predetermined performance deterioration is observed, based on the individual performance difference. Disk array controller. 5. The disk array control device according to claim 3, further comprising means for recording an error occurrence rate of each drive and selecting a drive based on the recorded error occurrence rate. 6. The disk array controller according to claim 3, further comprising means for allocating a dedicated area of the cache. 7. The disk array control device according to claim 3, further comprising means for controlling a disk cache based on the number of reduced logical volumes included in the disk array. 8. A control unit, a response time measuring unit, a cache control unit, and a cache, which are connected between a plurality of disks and a CPU via an interface, and which receives a Read / Write command from the CPU via the interface. A receiving disk array controller; an I / O interface and an I / O bus connected to the disk array controller;
e, a CPU that is a higher-level device that issues an instruction, wherein the response time measurement unit includes a counter, an average value calculation unit, a device table, a response time buffer, and a cache determination unit, and a Read / Write instruction from the control unit. In response to the issuance of the instruction, the counter calculates the time from issuance of an instruction to each device to a response, and inputs the response time of the instruction to the response time buffer. The average value of the response time up to and the number of instructions are stored. The average value calculation unit stores the average value of the response time up to the {m-1} th time and the number of instructions m− 1. From the response time of the m-th instruction held in the response time buffer, calculate the average value of the response time up to the m-th instruction, The cache determination unit determines cache allocation to each device based on the information in the device table, registers the device to be allocated in the device table, and the cache control unit 8. The disk array control device according to claim 3, wherein the control of the cache is performed according to cache allocation information included in the device table. 9. A disk array control method, comprising a step of allocating a part of cache capacity on a disk array exclusively to a device according to a response time of each device connected to the disk array device. 10. A performance difference between devices is determined by calculating an average value of response times of devices connected to a disk array within a certain time during actual operation, and comparing the average values. 10. The disk array control method according to claim 9, further comprising the step of: 11. The method according to claim 10, further comprising the step of allocating an arbitrary cache capacity exclusively to the disk having a predetermined performance difference based on the individual performance difference. Disk array control method. 12. The method according to claim 10, further comprising the step of allocating an arbitrary cache capacity exclusively to the disk on which a predetermined performance deterioration is observed, based on the individual performance difference. Disk array control method. 13. An error occurrence rate of each drive is recorded,
13. The method according to claim 11, further comprising a step of selecting a drive based on the recorded error occurrence rate.
3. The disk array control method according to 1. 14. The disk array control method according to claim 11, further comprising a step of allocating a dedicated area of the cache. 15. The disk array control method according to claim 11, further comprising the step of controlling a disk cache based on the number of reduced logical volumes included in the disk array.