JP2005025462A - Storage system, method and device for controlling variable control parameter of redundant control controller, and control program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To set the optimal variable control parameter for each of the applications of a plurality of hosts. <P>SOLUTION: This storage system is constituted of hosts 1-0 and so on and a disk array device 4 having a plurality of disk arrays equipped with logical units LU-0 and so on and a plurality of cooperative control controllers 5-1 and so on constituted by loosely coupling a plurality of tightly coupled redundant control controllers 5-1a and so on. The variable control parameters of the redundant control controllers 5-1a and so on are set according to the kinds of applications operating on the host 1-0, and when the parameter setting is changed, the variable control parameters set in the redundant control controllers 5-1a and so on are dynamically changed according to the parameter change. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a storage system having a plurality of logical units composed of a plurality of disk devices.
[0002]
[Prior art]
Conventionally, in a storage system having a plurality of logical units composed of a plurality of disk devices of this type, the disk device receives input / output requests from a plurality of applications operating on a plurality of hosts and follows the input / output requests via the control controller. I / O is being performed. The application mentioned here refers to an application that operates on a host such as a database, backup, streaming, or file server.
[0003]
At this time, the variable control parameters on the controller side are optimized according to the typical access pattern of input / output requests for multiple applications and the setting mode of clustering software running on the host side to improve availability. It is set to be.
[0004]
The variable control parameters generally have the following items.
[0005]
○ Logical unit unit
-RAID group Number of divisions / size
-RAID group stripe size
・ RAID level
○ Cache memory
-Write-back cache usage ratio
-Cache allocation ratio for sequential / random access
・ Cache discharge algorithm
○ Host interface
-Guaranteed bandwidth for each host interface port
-Number of command queuing per connected host
-Host I / F type
○ System
-Different failover methods for each clustering software used
(Reserve / release release method of logical unit by SCSI bus reset, reserve / release method to special logical unit, etc.)
For example, it is generally known that there are relatively many accesses with a relatively small block size in an application such as a database, and there are many accesses with a large block size in an application such as streaming. Such a typical operation pattern for each application has been set in the disk array in accordance with an analysis in the disk array or an instruction from the host side so as to optimize the setting of the variable control parameter.
[0006]
Also, there are cases where multiple disk units in one logical unit are shared by different hosts and different applications. In this case, variable control parameters are optimized for each disk unit corresponding to different hosts and different applications. Was set to be.
[0007]
Further, Patent Document 1 is known as a method for efficiently operating a storage system.
[0008]
[Patent Document 1]
JP 2003-134141 A
[0009]
[Problems to be solved by the invention]
As described above, the conventional technology optimizes the variable control parameters so that the variable control parameters of the disk devices corresponding to different hosts and different applications are optimized after allocating the resources of the entire system. It was. However, since optimization is performed by distributing the entire system resources in the disk array device, the optimization range is narrow, and optimization according to input / output requests from multiple hosts with different requirements and different applications can be achieved. There was a problem that it was difficult.
[0010]
Also, since different interface types cannot be supported at the same time, it is necessary to prepare a plurality of disk array devices and to correspond one logical unit to each host when connecting to hosts of different interface types.
[0011]
When trying to support clustering software on the host side, it is necessary to set variable control parameters related to the entire system for each clustering software with different control, so only one type of clustering software per system is required. Could not support. For this reason, when the number of hosts connected to a logical unit increases and support for multiple clustering software is required, it is necessary to prepare multiple logical units and to associate the logical unit with each clustering software. It was.
[0012]
In order to solve the above-mentioned problems, it is also possible to control by optimizing the variable control parameters of the entire system by sharing the disk device and the buffer memory in the disk device with two or more independent control controllers. However, it is necessary to configure the buffer memory and the disk device from such an independent controller as a transparent or limited buffer memory that appears to be transparent, which makes the configuration complicated and expensive. It was. As for the number of control controllers, there is a problem that the configuration is further complicated in order to obtain a redundant configuration of three or more units as compared to a redundant configuration of up to two units.
[0013]
In addition, by sharing a plurality of disk devices from a plurality of redundant control controllers, it may be realized by simplifying the control as compared with the case of controlling by two or more independent control controllers. . When a plurality of disk devices are shared by a plurality of redundant control controllers in this way, the bandwidth of the channel network / channel interface to which the plurality of disk devices are connected is also shared. There is a problem that the control parameter is optimally distributed among a plurality of redundant control controllers.
[0014]
It is an object of the present invention to provide a storage system, a variable control parameter control method and apparatus for a redundant control controller, and a control program that can set an optimal variable control parameter for each application of a plurality of hosts.
[0015]
[Means for Solving the Problems]
In order to achieve the above object, the present invention provides:
Multiple hosts,
A disk array having a plurality of logical units including at least one disk device, and a plurality of loosely coupled cooperative control controllers for controlling the disk array, wherein the plurality of cooperative control controllers are channels on the disk array side of each other. A plurality of tightly-coupled redundant control controllers connected by a network and connected to each other by a bus, and configured to control different logical units with variable control parameters, and data input / output with the plurality of hosts The disk array device
In a storage system with
Means for setting variable control parameters of the redundant controller according to the type of application running on the plurality of hosts;
Control means for dynamically changing a variable control parameter set in the redundant control controller according to the change when the parameter setting is changed;
It is characterized by comprising.
[0016]
In order to achieve the above object, the present invention provides:
Multiple hosts,
A disk array having a plurality of logical units including at least one disk device, and a plurality of loosely coupled cooperative control controllers for controlling the disk array, wherein the plurality of cooperative control controllers are channels on the disk array side of each other. A plurality of tightly-coupled redundant control controllers connected by a network and connected to each other by a bus, and configured to control different logical units with variable control parameters, and data input / output with the plurality of hosts The disk array device
In a method for controlling a variable control parameter of the redundant controller in a storage system comprising:
The variable control parameter of the redundant control controller is set according to the type of application running on the plurality of hosts, and when this parameter setting is changed, the variable control parameter set in the redundant control controller according to the change is set. It is characterized by changing dynamically.
[0017]
In order to achieve the above object, the present invention provides:
Multiple hosts,
A disk array having a plurality of logical units including at least one disk device, and a plurality of loosely coupled cooperative control controllers for controlling the disk array, wherein the plurality of cooperative control controllers are channels on the disk array side of each other. A plurality of tightly-coupled redundant control controllers connected by a network and connected to each other by a bus, and configured to control different logical units with variable control parameters, and data input / output with the plurality of hosts The disk array device
In the disk array device of the storage system comprising:
Means for setting variable control parameters of the redundant controller according to the type of application running on the plurality of hosts;
Control means for dynamically changing a variable control parameter set in the redundant control controller according to the change when the parameter setting is changed;
It is characterized by comprising.
[0018]
In order to achieve the above object, a storage system control program of the present invention provides:
Multiple hosts,
A disk array having a plurality of logical units including at least one disk device, and a plurality of loosely coupled cooperative control controllers for controlling the disk array, wherein the plurality of cooperative control controllers are channels on the disk array side of each other. A plurality of tightly-coupled redundant control controllers connected by a network and connected to each other by a bus, and configured to control different logical units with variable control parameters, and data input / output with the plurality of hosts The disk array device
For computers applied to storage systems with
A procedure for setting the variable control parameter of the redundant controller according to the type of application running on the plurality of hosts;
A procedure for dynamically changing a variable control parameter set in the redundant controller according to the change when the parameter setting is changed;
Is executed.
[0019]
According to the storage system, the control method and apparatus for the variable control parameter of the redundant control controller, and the control program of the present invention, the variable control parameter of the redundant control controller can be dynamically changed according to the type of application. It is possible to set an optimum variable control parameter for each host application.
[0020]
More specifically, redistribution is performed for each of the plurality of redundant control controllers so that the resource allocation becomes optimal when the types are divided. For example, control is performed so that the volume of the combination of sequential access and sequential access patterns decreases, or the combination of sequential and random increases.
[0021]
Also, when using a shared disk with raster software, etc., by using a volume that forgets the reserve information by issuing a reset via the channel / network, and by using some special reserve via the channel / network It is possible to support a plurality of clustering software by setting the volume so that the reserve information is not forgotten just by issuing a reset.
[0022]
Further, for each control controller, this control controller performs parameter tuning dedicated to Oracle, and a certain controller performs parameter tuning dedicated to data backup. As a result, it is possible to optimally tune parameters of a plurality of host devices and a plurality of application disk devices within one disk array device.
[0023]
In addition, by specifying the volume type at the time of takeover, for example, the mirror volume in the device uses a large cache area to perform the copy operation asynchronously and at high speed, and backs up the volume type when detached Switching to the dedicated type, the cache usage is small, and the HDD performs operations such as sequential access and bandwidth limitation separately.
[0024]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, a storage system according to the present invention will be described with reference to the drawings.
[0025]
(First embodiment)
FIG. 1 is a diagram showing a storage system according to the first embodiment of the present invention.
[0026]
As shown in FIG. 1, the storage system of this embodiment includes hosts 1-0 to 1-M, a network / channel network 3, and a disk array device 4.
[0027]
The hosts 1-0 to 1-M are connected to the disk array device 4 via the network / channel network 3, and a plurality of hosts share one disk array device 4. Hosts 1-0 to 1-M are indicated by identification numbers # 0 to #M.
[0028]
Each of the hosts 1-0 to 1-M has an application. For example, the host 1-0 has an application 2-1. The applications referred to here are database applications, clustering software applications, backup applications, mail applications, streaming applications, and the like.
[0029]
The disk array device 4 includes loosely coupled cooperative control controllers 5-1 and 5-2, a channel network 6, and a disk array 7 including logical units LU-m to LU-0 made up of a large number of disk devices. ing. Here, the logical units LU-m to LU-0 are indicated by identification numbers # 0 to #M.
[0030]
The cooperative control controllers 5-1 and 5-2 control the disk array 7 and are connected to the disk array 7 via the channel network 6. The cooperation control controllers 5-1 and 5-2 are connected to each other via the channel network 6.
[0031]
The cooperative control controller 5-1 includes closely-coupled redundant control controllers 5-1a and 5-1b that are connected to each other via a bus. Similarly, the cooperative control controller 5-2 has redundant control controllers 5-2a and 5-2b connected to each other via a bus.
[0032]
Each redundant controller 5-1a, 5-1b, 5-2a, 5-2b has a disk control unit 8, a buffer memory 9, a host I / F unit 10, an HDD I / F unit 12, and an inter-controller communication unit 12. These are connected to each other via an internal bus 11. Each of these redundant control controllers 5-1a, 5-1b, 5-2a, 5-2b has variable control parameters, and these parameters are held in the management tables shown in FIGS.
[0033]
FIG. 2 shows a parameter management table, and the management tables {circle around (1)} to {circle around (4)} held by the redundant control controller respectively hold parameters. Here, the variable control parameter has the same information in the two redundant control controllers made redundant, for example, the redundant control controller 5-1a and the redundant control controller 5-1b.
[0034]
The management table indicated by {circle over (1)} holds a single parameter for each redundant control controller, and the parameter is, for example, a system identification number or a logical unit reserve / release control method.
[0035]
The management table indicated by {circle over (2)} holds parameters for each logical unit in the redundant control controller. Examples of the parameters are shown below.
[0036]
・ RAID division number / size
・ RAID stripe size
-RAID level
The management table indicated by {circle around (3)} holds parameters related to the buffer memory in the redundant control controller. Examples of the parameters are shown below.
[0037]
-Write-back cache usage ratio
-Cache allocation ratio for sequential / random access
・ Cache discharge algorithm
-Cache usage for each host-LU path
The management table indicated by {circle over (4)} holds parameters for each host I / F of the redundant control controller. Examples of the parameters are shown below.
[0038]
-Bandwidth allocation for each host interface port
-Number of command queuing per connected host
-Host I / F type
FIG. 3 shows a redundant control controller management table, which shows combinations of logical units, redundant control controllers, and parameters shown in FIG.
[0039]
For example, in the case of correspondence between the logical unit of number # 0 and the redundant controller of number # 0, four types (1) to (4) are defined as management parameters. For example, type {circle around (1)} indicates parameter {circle around (1)}, that is, a parameter that is uniquely held for each redundant controller in FIG.
[0040]
The disk controller 8 controls the entire disk array 7 and writes to the logical units LU-m to LU-0 assigned to the disk array 7 in response to a command from the host 1-0 or the like. / Read control. In addition, takeover processing is performed based on a delivery instruction or a takeover instruction from the host 1-0 or the like.
[0041]
The cache memory 9 includes a tag part and a data part. The tag part stores the address, size, attribute information, pointer, etc. of the logical unit data stored in the data part. Here, the attribute information is information indicating validity / invalidity.
[0042]
The host I / F unit 10 is an interface that accepts an input / output request from the host 1-0 and the like.
[0043]
The HDD I / F unit 12 is an interface with the disk array 7 connected to the channel network 6.
[0044]
The inter-controller communication unit 12 is used for high-speed data / control information communication between the redundant control controllers 5-1a, 5-1b, 5-2a, and 5-2b.
[0045]
Next, the operation of the storage system of this embodiment will be described with reference to FIG.
[0046]
First, when the host control unit 14 detects that the management table of the redundant control controller #N has been updated by, for example, polling or the like, the changed parameter is specified in the parameter change table (step S1).
[0047]
Next, the host control unit 14 instructs the redundant control controller that has been changed as indicated by reference numerals 100 and 101 in FIG. The parameter change instruction at this time is made via the network / channel network, specifically by a SCSI command or a control frame on TCP / IP (step S2).
[0048]
Next, the redundant controller #N receives the parameter change instruction via the host I / F (step S3).
[0049]
Next, according to the changed parameter, the redundant control controller #N performs matching of information in the same redundant control controller and parameter change in the same redundant control controller (step S4).
[0050]
Next, the host control unit 14 checks whether or not the management table has been updated by another redundant control controller, and if it has been updated, returns to the procedure (1). If not updated otherwise, the process ends (step S5).
[0051]
As described above, the storage system of this embodiment includes a plurality of hosts 1-0... And a loosely coupled cooperative controller 5-1 that performs control on a disk array LU having a plurality of logical units. In the storage system including the disk array device 4 having a plurality of redundantly coupled redundant controllers 5-1a ..., the controllers 5-1 ... are newly provided with host controllers 14 on the hosts 1-0 ... The variable control parameters of the redundant controller 5-1a... Are set according to the type of application to be operated, and when the parameter setting is changed, the variable control parameter set in the redundant controller 5-1a. Since it changes dynamically, the redundant control controller 5 depends on the type of application. Since 1a ... variable control parameters can be changed dynamically, it is possible to set the optimum variable control parameter for each of the plurality of host 1-0 ... applications.
[0052]
(Second Embodiment)
Next, a second embodiment of the storage system according to the present invention will be described with reference to FIGS.
[0053]
The overall system configuration of this embodiment is almost the same as that in FIG. 1, and the management table is different. That is, FIG. 5 shows a Nexus management table, that is, a management table of host-application-redundant control controller relationships, which correspond to logical unit numbers, redundant control controller numbers, and corresponding host numbers. The relationship between the application running on the host and the management parameters is shown. For example, the logical unit # 0 and the redundant controller # 0 correspond to the host # 0 and the database as an application operating on the host # 0, and the management parameters in this correspondence relationship are shown in FIG. Four types (1) to (4) are defined. For example, type {circle around (1)} indicates parameter {circle around (1)}, that is, a parameter that is uniquely held for each redundant controller in FIG.
[0054]
FIG. 6 shows a management parameter table for each application, and shows the relationship between the application running on the corresponding host and the management parameter. For example, the four types (1) to (4) shown in FIG. 2 are defined as management parameters when the application is a database. For example, type {circle around (1)} indicates parameter {circle around (1)}, that is, a parameter that is uniquely held for each redundant controller in FIG.
[0055]
Next, the operation of the storage system of this embodiment will be described with reference to FIG.
[0056]
First, the host and the application running on the host are changed (step T1).
[0057]
Next, the host control unit 14 searches the Nexus management table using the host as a key (step T2).
[0058]
Next, if the corresponding entry exists, the process proceeds to step T4. If not, the process proceeds to step T11 (step T3).
[0059]
Next, it is checked whether there is a change in the application of the corresponding entry. If there is a change, the process proceeds to step T5, and if not, the process proceeds to step T10.
[0060]
In step T5, a variable control parameter suitable for the application designated from the application-specific management table shown in FIG. 6 is selected.
[0061]
Next, the host controller 14 instructs the redundant controller #N to change parameters (step T6).
[0062]
Next, the redundant controller #N receives parameters (step T7).
[0063]
Next, the parameter is changed in the redundant control controller # (step T8).
[0064]
Next, the Nexus management table is updated (step T9).
[0065]
If the host / application relationship is to be changed, the process returns to step T1. If not changed, the process ends (step T10).
[0066]
If the corresponding entry does not exist in step T3 and the host is not yet registered, the host controller 14 is instructed to the redundant control controller corresponding to the host and the application, and the process proceeds to step T5.
[0067]
As described above, the storage system according to the present embodiment includes a plurality of hosts 1-0, and a loosely coupled cooperative controller 5-1 that performs control on a disk array LU having a plurality of logical units. 5-1... Are newly provided with the host control unit 14 in the storage system including the disk array apparatus 4 having a plurality of redundantly coupled redundant control controllers 5-1a. Since the corresponding application is instructed and the variable control parameter of the redundant controller 5-1a... Is converted into the variable control parameter corresponding to the instructed application, it is optimal for each application of the plurality of hosts 1-0. Variable control parameters can be set.
[0068]
(Third embodiment)
Next, a third embodiment of the storage system according to the present invention will be described with reference to FIGS.
[0069]
As shown in FIG. 8, the storage system of this embodiment includes a host control unit 14 'different from that shown in FIG. The host control unit 14 'relates to statistical information on processing according to an input / output request from an application for each redundant control controller 5-1a ... by periodic polling communication or the like and occurrence of a failure for each redundant control controller 5-1a ... Statistical information is collected, it is determined whether or not the value of the collected statistical information exceeds a preset threshold value, and the application allocation is reviewed for each redundant control controller 5-1a... The update processing unit 14'A for resetting the variable control parameters of each redundant control controller 5-1a ... based on the allocation result. A specific variable control parameter update process is performed by a Nexus management table which is a management table of a host-application-redundant control controller relationship.
[0070]
FIG. 9 shows the Nexus management table of this embodiment, and the management table shows the relationship of host-application-redundant control controller. For example, the management parameters when the logical disk # 0, the redundant controller # 0, the host # 0, and the application running on the host # 0 are databases are defined as four types (1) to (4) shown in FIG. Has been. For example, type {circle around (1)} indicates parameter {circle around (1)}, that is, a parameter that is uniquely held for each redundant controller in FIG.
[0071]
FIG. 10 shows a table that holds statistical information collected from each redundant control controller in all the disk array devices 4, and holds management parameters and threshold values by a combination of the disk array number and the redundant control controller number. ing. For example, in the combination of the disk array # 0 and the redundant controller # 0, the statistical information groups (1) to (4) are held, and the thresholds (1) to (4) are associated with these statistical information groups (1) to (4). It has (4).
[0072]
Next, the operation of the storage system of this embodiment will be described with reference to FIG. The operation of the storage system of this embodiment shown in FIG. 11 is obtained by replacing step T1 in FIG. 7 with steps T12 to T15.
[0073]
In step T12 in FIG. 11, the host control unit 14 ′ collects statistical information from each redundant controller 5-1a... By an arbitrary method such as periodic polling.
[0074]
Next, the host control unit 14 'updates the statistical information table and compares it with a preset threshold value (step T13).
[0075]
Next, when the statistical information content exceeds a preset threshold value, for example, when the response time is slow, refer to the statistical information table from the entries registered in the Nexus management table of FIG. Then, an entry having a sufficient processing capacity is selected (step T14).
[0076]
Next, among all logical units controlled by the redundant controller 5-1a... Exceeding the threshold, it is decided to move the control of the logical unit exceeding the threshold to an entry that still has sufficient processing capacity. (Step T15).
[0077]
The subsequent steps T2 to T11 are the same as the operation of the second embodiment shown in FIG.
[0078]
In this embodiment, statistical information is collected by the host control unit 14 'and the variable control parameters of each redundant controller 5-1a ... are reset as in the second embodiment, but this is particularly suitable in the following cases. For example, the disk array # 0 and the redundant controller have been online (to the database), but if the access time of the index section exceeds a predetermined time, only the disk array that constitutes the database index The parameter is reset so that the control is transferred to another redundant control controller with a relatively small load, and it is possible to avoid an increase in processing time by using the entire buffer memory.
[0079]
In addition, when the HDD access time increases due to the increase in the number of HDD retries, the cache area decreases due to the cache memory degeneration, and the write back cache area degenerates due to the decrease in the battery charge amount, etc. Such a problem is solved by resetting the parameters of this embodiment.
[0080]
(Fourth embodiment)
Next, a fourth embodiment of the storage system according to the present invention will be described with reference to FIGS.
[0081]
The storage system of this embodiment includes a host control unit 14 'shown in FIG. 8 and collects general information, but its table configuration and processing are different from those of the third embodiment.
[0082]
As shown in FIG. 12, the Nexus management table (master), which is a management table of the host-application-redundant control controller relationship, has a host # 0 corresponding to the disk array # 0 and the redundant control controller # 0. As the management parameters when the application operating on the host is a database application, the four types {circle around (1)} to {circle around (4)} shown in FIG. 2 are defined. For example, type {circle around (1)} indicates parameter {circle around (1)}, that is, a parameter that is uniquely held for each redundant controller in FIG.
[0083]
FIG. 13 shows a schedule table collected from each redundant control controller in all the disk array devices. For example, when there are two redundant control controllers, this table sets variable control parameters suitable for online operations such as a database during the daytime, performs online work with only one at night, and the remaining one It is assumed that backup will be performed at
[0084]
Next, the operation of the storage system of this embodiment will be described with reference to FIG. The operation of the storage system of this embodiment shown in FIG. 14 is obtained by replacing step T1 in FIG. 7 with steps T16 to T18.
[0085]
In step T16, the time table and the Nexus management table (# 1 to #N) shown in FIG. 12 corresponding to the time table are registered in the schedule table of the host control unit 14 ′.
[0086]
Next, the host control unit 14 'periodically checks the time table or detects that it is time to update the Nexus management table (master) by event notification (step T17).
[0087]
Next, the host control unit 14 ′ compares each entry in the Nexus management table (master) to be changed in the corresponding schedule management table entry shown in FIG. 12 with each entry in the Nexus management table (#xx). The different entry is detected, and the process for the different entry is the same as the process from step T2 to T11 of the second embodiment 2 (step T18).
[0088]
When there are two redundant control controllers as in this embodiment, variable control parameters suitable for online operations such as databases can be set for both units during the daytime, and online operations can be performed with only one at night, Backup can be easily performed with the remaining one.
[0089]
(Fifth embodiment)
Next, a fifth embodiment of the storage system according to the present invention will be described with reference to FIGS.
[0090]
As shown in FIG. 15, the storage system of this embodiment includes a host control unit 14 'different from that shown in FIGS. 1 and 8, and the other configuration is the same as that shown in FIGS. The host control unit 14 ″ has an optimization unit 14 ″ A. This optimization unit 14 ″ A performs optimization by taking over variable control parameters when each logical unit LU-1... Managed by one redundant control controller is taken over by another redundant control controller. is there.
[0091]
A specific variable control parameter update process is performed by a Nexus management table which is a management table of a host-application-redundant control controller relationship.
[0092]
FIG. 16 shows the Nexus management table of this embodiment, and this management table shows the relationship of host-application-redundant control controller. For example, the management parameters in the case where the disk array # 0, the redundant controller # 0, the logical units LU-0 and LU-1, the host # 0, and the application operating on the host # 0 are databases are shown in FIG. Types (1) to (4) are defined. For example, type {circle around (1)} indicates parameter {circle around (1)}, that is, a parameter that is uniquely held for each redundant controller in FIG.
[0093]
Next, the operation of the storage system of this embodiment will be described with reference to FIG. The operation of the storage system of this embodiment shown in FIG. 17 is obtained by replacing step T1 in FIG. 7 with steps T19 to T22.
[0094]
In step T19, the host controller 14 ″ is instructed to take over the logical unit LU-1 or LU-0 and LU-1 controlled by the redundant controller # 0 of the disk array # 0.
[0095]
Next, the host control unit 14 ″ selects an entry including the redundant control controller number of the disk array including the corresponding logical unit from the Nexus management table of FIG. 16 (step T20).
[0096]
Next, the host control unit 14 ″ selects a parameter to be taken over from the corresponding host, the corresponding application, and the management parameter (step T21).
[0097]
Next, the host control unit 14 ″ decides to move the corresponding host, the corresponding application, and the management parameter to be taken over to the entry of the redundancy controller # 1, and for the entry with the change, the process The same processing as that in steps T2 to T11 of the second embodiment is performed (step T22).
[0098]
As described above, according to this embodiment, when each logical unit LU-1... Managed by one redundant control controller is taken over by another redundant control controller, optimization can be performed by taking over variable control parameters. It becomes possible.
[0099]
The configuration of the storage system according to the present invention described above and illustrated is that the control system excluding the disk array can be configured by one or a plurality of computers or a part of the functions of these computers, and these hardware configurations are configured by the computer. It can also be configured as a program to be executed.
[0100]
【The invention's effect】
As described above, according to the present invention, the control variable parameter in the disk array in which a plurality of redundant control controllers are linked can be changed greatly, so that it is optimal for each host application connected to the redundant control controller. It is possible to provide a storage system capable of setting various control variable parameters, a control method and apparatus for variable control parameters of a redundant control controller, and a control program.
[Brief description of the drawings]
FIG. 1 is a configuration diagram of a storage system according to a first embodiment of the invention.
FIG. 2 is a view showing a parameter management table in the embodiment.
FIG. 3 is a view showing a redundant control controller management table in the same embodiment;
FIG. 4 is a flowchart showing the operation of the embodiment.
FIG. 5 is a diagram showing a Nexus management table according to the second embodiment of the present invention.
FIG. 6 is a view showing a management parameter table for each application in the embodiment.
FIG. 7 is a flowchart showing the operation of the embodiment.
FIG. 8 is a configuration diagram of a storage system according to a third embodiment of the invention.
FIG. 9 is a view showing a Nexus management table in the embodiment;
FIG. 10 is a view showing a statistical information table in the embodiment.
FIG. 11 is a flowchart showing the operation of the embodiment.
FIG. 12 is a diagram showing a Nexus management table according to the fourth embodiment of the present invention.
FIG. 13 is a view showing a schedule table according to the embodiment;
FIG. 14 is a flowchart showing the operation of the embodiment.
FIG. 15 is a configuration diagram of a storage system according to a fifth embodiment of the invention.
FIG. 16 is a view showing a Nexus management table in the embodiment;
FIG. 17 is a flowchart showing the operation of the embodiment.
[Explanation of symbols]
1-0 to 1-M: Host, 2 ... Application, 3 ... Network / Channel network, 4 ... Disk array device, 5-1, 5-2 ... Cooperation controller, 5-1a, 5-1b, 5- 2a, 5-2b ... redundant controller, 7 ... disk array, 8 ... disk controller, 9 ... buffer memory, 10 ... host I / F unit, 11 ... internal bus, 12 ... HDD I / F unit, 13 ... between controllers Communication unit 14, 14 ', 14 "... host control unit.

Claims (9)

Multiple hosts,
A disk array having a plurality of logical units including at least one disk device, and a plurality of loosely coupled cooperative control controllers for controlling the disk array, wherein the plurality of cooperative control controllers are channels on the disk array side of each other. A plurality of tightly-coupled redundant control controllers connected by a network and connected to each other by a bus, and configured to control different logical units with variable control parameters, and data input / output with the plurality of hosts In a storage system equipped with a disk array device in which
Means for setting variable control parameters of the redundant controller according to the type of application running on the plurality of hosts;
A storage system comprising: control means for dynamically changing a variable control parameter set in the redundant control controller according to the change when the parameter setting is changed. Means for indicating an application corresponding to the redundant controller;
The storage system according to claim 1, further comprising means for converting a variable control parameter of the redundant control controller into a variable control parameter corresponding to the instructed application. Means for collecting statistical information on processing according to an input / output request from an application for each redundant control controller and statistical information on occurrence of a failure for each redundant control controller;
Means for determining whether or not the value of the collected statistical information exceeds a preset threshold;
Means for reviewing application allocation for each redundant control controller from the result determined from statistical information;
3. The storage system according to claim 2, further comprising means for resetting variable control parameters of each redundant controller based on the reviewed allocation result. Means for holding an application allocation schedule for each redundant control controller;
The storage system according to claim 1, further comprising means for resetting variable control parameters of the redundant controller based on an allocation schedule according to the schedule. Means for taking over the variable control parameter information when taking over control of the disk array from one redundant controller to another redundant controller in the plurality of redundant controllers;
2. The storage system according to claim 1, further comprising means for redistributing variable control parameters corresponding to the disk array to which control is to be taken over in the takeover destination redundant control controller. The redundant controller is
A host interface that accepts an I / O request from the host; receives an I / O request from the host; analyzes the accepted content; outputs a request to the disk array corresponding to the request from the host; Between the control unit that performs control to return the data to the host, the buffer memory that includes tag information for accessing the data of the logical unit according to the request at high speed, and the disk devices that constitute the disk array 6. A disk device interface unit for making a request and an inter-controller communication control unit for performing high-speed data / control information communication between the redundant control controllers. The described storage system. Multiple hosts,
A disk array having a plurality of logical units including at least one disk device, and a plurality of loosely coupled cooperative control controllers for controlling the disk array, wherein the plurality of cooperative control controllers are channels on the disk array side of each other. A plurality of tightly-coupled redundant control controllers connected by a network and connected to each other by a bus, and configured to control different logical units with variable control parameters, and data input / output with the plurality of hosts In the method of controlling the variable control parameter of the redundant controller in a storage system comprising a disk array device in which
The variable control parameter of the redundant control controller is set according to the type of application running on the plurality of hosts, and when this parameter setting is changed, the variable control parameter set in the redundant control controller according to the change is set. A method for controlling a variable control parameter in a storage system, wherein the variable control parameter is dynamically changed. Multiple hosts,
A disk array having a plurality of logical units including at least one disk device, and a plurality of loosely coupled cooperative control controllers for controlling the disk array, wherein the plurality of cooperative control controllers are channels on the disk array side of each other. A plurality of tightly-coupled redundant control controllers connected by a network and connected to each other by a bus, and configured to control different logical units with variable control parameters, and data input / output with the plurality of hosts In the disk array device of the storage system comprising the disk array device in which
Means for setting variable control parameters of the redundant controller according to the type of application running on the plurality of hosts;
A disk array device for a storage system, comprising: control means for dynamically changing a variable control parameter set in the redundant controller according to the change of parameter setting. Multiple hosts,
A disk array having a plurality of logical units including at least one disk device, and a plurality of loosely coupled cooperative control controllers for controlling the disk array, wherein the plurality of cooperative control controllers are channels on the disk array side of each other. A plurality of tightly-coupled redundant control controllers connected by a network and connected to each other by a bus, and configured to control different logical units with variable control parameters, and data input / output with the plurality of hosts To a computer applied to a storage system having a disk array device
A procedure for setting the variable control parameter of the redundant controller according to the type of application running on the plurality of hosts;
A control program for a storage system for executing a procedure for dynamically changing a variable control parameter set in the redundant control controller according to the change when the parameter setting is changed.

Images