Data mapping management apparatus

The present invention relates to a computer system in which devices and software having a function of providing a virtual storage area to a computer are combined.

PROBLEM TO BE SOLVED: To increase the storage capacity of a storage device of a computer system as the amount of data stored in the computer system increases. In order to reduce the introduction, maintenance, and management costs of large-capacity storage devices, storage devices have come to be centralized using SAN (Storage Area Network), which is a network dedicated to storage.

[0003] In a SAN environment, since a plurality of computers and storage devices are connected to one network, it is necessary to manage the correspondence between the computers and the storage devices, access restrictions for specific data, and the like. In addition, when constructing a SAN, a network using a switch represented by FC (Fibre Channel) is often used. The Sui the SAN of Tchibesu is large-scale, network using a plurality of switches is formed, it is necessary to manage the respective switches.

[0004] In order to facilitate the management of a plurality of switches, a management function that integrates management, particularly a management function that uses a GUI (Graphical User Interface) is used. In "Storage Management Solution in SAN Environment" (Igarashi Isao, FUJITSU 2001-March issue, Vol.52, No.2, Vol. 303, pp.118-122, Fujitsu Limited, ISSN 0016-2515) The management software that integrates configuration management in SAN using FC, access path management including device allocation, failure management, backup management, etc. is explained.

[0005] One of the important things in the management of a computer system is the management of data correspondence (hereinafter, "data mapping"). US Pat. No. 6,035,306 discloses a technology related to a performance analysis tool for realizing data mapping optimization from the viewpoint of performance in a computer system that executes a DBMS for the purpose of improving the efficiency of data mapping management. In this technology, load information of objects in each layer considering data mapping between DBMS, file system, volume manager, and storage device is displayed on the screen. In addition, a function of displaying a screen showing load information of an object of another hierarchy corresponding to an object of a certain hierarchy is provided.

[0006] On the other hand, for the purpose of efficient use of the storage device in the storage system in the SAN environment and facilitation of storage system management, the physical storage device for storing data is not a part of the physical storage device as it is, or a part of the physical storage device or There is a technology called "virtualization" that creates a virtual storage area by combining all the storage areas and provides it to a computer or an application program executed by the computer.

[0007] Virtualization includes a function of combining a plurality of physical storage devices into one virtual storage area, a function of dividing one physical storage device into a plurality of virtual storage areas, and the like.

[0008] Further, Japanese Patent Application Laid-Open No. 9-274544 provides a logical storage device for a computer, and in a storage device that maps and stores the data in an area on the physical storage device, the data storage position is described. A technique for improving the access performance of a storage device by providing a dynamic change function and optimizing the physical storage position of data is disclosed.

[0009] Japanese Unexamined Patent Publication No. 2000-339098 discloses a technique for providing a virtualization function in a device called a storage domain management system existing between a computer and an I / O path in a storage device. This device provides a computer with a logical storage range called a storage domain, a logical unit of a virtualized storage device, and the like. The data is stored in the physical storage area of the storage device recognized by the device. Then, this device executes data read / write processing to the corresponding physical area in response to an access request from the computer. In addition, the storage domain management system provides the ability to dynamically change the data storage location.

[0010] [Problems to be Solved by the Invention] The above-mentioned virtualization function can be introduced into all devices related to I / O execution from a computer to a storage device. Therefore, it is possible to construct a computer system having a virtualization function with a plurality of devices. In this case, the user will use the function of providing the virtual storage area hierarchically.

[0011] When changing the data mapping in a computer system having a virtualization function, the user grasps which device provides the data mapping change function, and then virtualizes the storage area in each device. It is necessary to grasp the data mapping by, determine which device and how to change the data mapping, create a detailed processing procedure, and then issue a data mapping change instruction.

[0012] As described above, the prior art provides a configuration management screen using a GUI, an operation information display function in consideration of data mapping, and a dynamic change function of a data storage position. In a simple combination of these prior arts, the administrator must consider the details of the processing procedure to achieve the data mapping desired by the administrator. In some cases, it is necessary not only to instruct the data storage position to be changed, but also to create a new virtual storage area or change the configuration of the virtual storage area. However, creating these processing procedures to achieve purposeful data mapping while considering multi-layered mapping is especially important for large-scale computer systems or when certain data is shared by multiple computers. It is not easy when there is.

[0013] An object of the present invention is to provide a virtual storage area, that is, in a computer system in which there are a plurality of devices that provide virtualization, an administrator can easily change the data mapping performed by the virtualization in each device. Is to provide a way to do it.

[0014] A further object of the present invention is to provide a method for realizing efficient data mapping in a computer system having a multi-layered virtual storage area.

[Means for Solving the Problems] The following means are provided for achieving the object of the present invention.

[0016] A data mapping management server collects data mapping information in a device that provides a virtual storage area in a computer system, for example, a switch that is a data transfer device, and virtual storage is based on the collected information. It provides the administrator with a management screen that can comprehensively grasp the virtual storage area created and managed in each device that provides the area. The screen is used to accept instructions for changing the data mapping. The data mapping management server creates a draft processing procedure for realizing the mapping change instruction of the received data, and presents it to the administrator. The administrator decides whether or not the plan can be executed, and if the administrator permits the execution, the virtual storage area providing device is instructed to move data or create a new virtual storage area according to the creation plan. Issuance.

[0017] When receiving a change instruction of data mapping, it provides a function of instructing data whose mapping should be changed by a virtual storage area in an arbitrary virtual storage area providing device or a physical storage area in the storage device. In addition, the virtual storage area providing device and the storage device, and the virtual storage area and the physical storage area provided by them can be specified as the data mapping change destination.

[0018] Provide a function of creating a procedure plan for processing a data mapping change, a virtual area creation, or a configuration change, which is necessary when the data mapping instructed by the administrator is realized in the computer system. At this time, the data mapping change processing procedure plan to efficiently use the data mapping change function provided by multiple devices by grasping all the areas that can be used as the mapping change destination and reusing them as necessary. create. In addition, check whether the data to be changed by mapping is shared by multiple computers, and if it is shared, create a data mapping change processing procedure plan so that data sharing can be continued without problems during or after the data mapping is changed. To do.

[0019] Further, the data mapping management server collects load information of the virtual storage area in the device that provides the virtual storage area or the physical storage area in the storage device. When creating a data mapping change processing procedure plan, create a processing procedure plan that takes load information into consideration, and realize data mapping in which problems related to access performance are unlikely to occur.

[Embodiment of the Invention] FIG. 1 is a diagram showing a configuration of an embodiment of a computer system to which the present invention is applied. The computer system includes a computer that uses a storage device (hereinafter, "host") 70, a computer that manages virtualization (hereinafter, "data mapping management server") 72, a virtualization switch 60 that performs virtualization processing, and a storage device 50. Has. Each device is connected to network 46 and can communicate with each other.

[0021] The host 70, the virtualization switch 60, and the storage device 50 are connected to each other by a communication line (hereinafter, I / O path) 40. The I / O processing between the host 70 and the storage device 50 is performed using the I / O path 40. The I / O path 40 may use a communication line that transfers data between devices using different physical media or different protocols. Further, the network 46 and the I / O path 40 may be the same communication line.

The storage device 50 includes a control device 52 and a disk device 54. These devices are connected by an internal bus 56. The control device 52 has a management unit 104a, mapping information 10a, and load information 12a. The disk device 54 may be singular or plural.

The storage device 50 virtualizes the physical storage area of the disk device 54 and provides one or more logical disk devices 208 to the external device. The provided logical disk device 208 may have a one-to-one correspondence with the disk device 54, or may correspond to a storage area composed of a plurality of disk devices 54. The management unit 104a controls the address translation between the logical disk device 208 and the physical storage area of the disk device 54 by using the mapping information 10a.

[0024] The mapping information 10a is information indicating a correspondence relationship between the logical disk device 208 and the physical storage area of the disk device 54. The management unit 104a measures the load applied to a specific logical disk device 208 or disk device 54. The load information 12a is information indicating the load measured by the management unit 104a. Further, the storage device 50 transmits the mapping information 10a and the load information 12a to other devices through the network 46.

Further, the management unit 104a dynamically changes the storage position of the data belonging to the logical disk device 208 in the disk device 54 according to the instruction received from the outside via the network 46. "Dynamic change" means "change the storage position without interrupting system processing". At this time, the management unit 104a can execute both the change of the data storage position accompanied by the copy of the data and the update process of only the mapping information 10a without the copy of the data.

[0026] Note that the management unit 104a of the storage device 50 does not necessarily have to collect the load information 12a. Further, the management unit 104a does not necessarily have to dynamically change the data storage position. Further, the management unit 104a may not be able to execute the change process of only the mapping information 10a without copying the data.

[0027] The virtualization switch 60 has a control unit 62. The control unit 62 has a management unit 104b, mapping information 10b, and load information 12b. The virtualization switch 60 recognizes the logical disk device 208 provided by the device connected to the virtualization switch 60 by using the management unit 104b, and virtualizes the storage area of the recognized logical disk device 208. Provide 206 to other devices. When the virtualization switch 60 is connected to another virtualization switch 60, the virtualization switch 60 recognizes the virtual volume 206 provided by the other virtualized switch 60 to be connected in addition to the logical disk device 208. Then, the recognized virtual volume 206 and the logical disk device 208 are further virtualized to provide the virtual volume 206 to other devices.

[0028] The mapping information 10b stores information indicating the correspondence between the virtual volume 206 and the logical disk device 208 (or the logical disk device 208 and the virtual volume 206). In addition, the management unit 104b measures the load related to the virtual volume 206 from the logical disk device 208 or the other virtualization switch 60, which is a component of the virtual volume 206 provided by the virtualization switch 60 to other devices. The load information 12b is information indicating the measured load. Further, the virtualization switch 60 transfers the mapping information 10b and the load information 12b to another device via the network 46.

[0029] Further, the management unit 104b dynamically changes the storage position of the component of the virtual volume 206 provided by the virtualization switch 60 based on an instruction from another device. At this time, the management unit 104 can perform both the data movement accompanied by the copy of the data and the update process of only the mapping information 10b without the copy of the data.

[0030] The virtualization switch 60 does not necessarily have to collect the load information 12b. Further, the virtualization switch 60 does not necessarily have to dynamically change the data storage position. Further, the virtualization switch 60 does not have to change the mapping information 10b without copying the data.

The host 70 has a CPU and memory, and the CPU executes programs such as an operating system (OS) 80, an application 88, and a management agent program 102. OS80 includes Volume Manager 84 and File System 86. Further, the OS 80 may have a raw device mechanism that enables direct access to a storage area such as a virtual volume 206 with the same file I / O interface as the software interface for the file 202 of the application program 88.

[0032] In this figure, the host 70 has only one file system 86, but the host 70 may have a plurality of file systems 70. The application program 88 is a program that uses the data stored in the storage area provided by the storage device 50.

[0033] The volume manager 84 recognizes a virtual volume 206 or a logical disk device 208 provided by another device by the host 70, and combines at least one virtual storage area belonging to the virtual volume 206 or the logical disk device 208. A program executed by the CPU of host 70 to provide logical volume 204 to file system 86 or application program 88. These programs are installed on host 70 using a network or storage medium.

[0034] The volume manager 84 has mapping information 10c which is information indicating a correspondence relationship between the logical volume 204 and the virtual volume 206 or the logical disk device 208. Volume manager 84 may make dynamic changes to the data storage location.

[0035] The file system 86 recognizes the logical volume 204, the virtual volume 206, or the logical disk device 208 by the host 70, extracts a part of the storage area belonging to them, and makes the file 202 easy to use as an application program. A program that runs on the CPU to provide to 88. Host 70 holds mapping information 10d for managing file 202. The file system 86 may have a function of dynamically changing the data storage position.

The storage device 50, the virtualization switch 60, the volume manager 84, and the file system 86 (hereinafter collectively referred to as virtualization mechanism) are the file 202, the logical volume 204, the virtual volume 206, and the logic, respectively. It has a function to dynamically create / delete a disk device 208 (hereinafter, these are collectively referred to as a "virtual structure", and a virtual structure plus a disk device 54 is collectively referred to as a "management structure"). .. Here, "dynamically creating / deleting" means creating / deleting a virtual structure without stopping the operation of the computer system.

[0037] When the virtualization mechanism has a function of dynamically changing the storage position of data and the function changes the mapping information 10a, 10b, 10c, and 10d without copying the data, the virtualization mechanism When updating the mapping information 10a, etc., changes that accompany the expansion / contraction of the storage area of the virtual structure are allowed. In addition, the virtualization mechanism has a function of suspending the I / O processing required by the host 70.

[0038] The management agent program 102 is executed when the host 70 acquires the mapping information 10c and 10d held by the volume manager 84 and the file system 86 and sends them to the data mapping management server 72 via the network 46. .. Also, when issuing instructions received from the data mapping management server 72 to the volume manager 84 or file system 86, such as dynamic change of data mapping, creation / deletion of virtual structure, suspension of I / O processing, etc. , This program is executed.

[0039] The data mapping management server 72 has a CPU and a memory. The data mapping management program 100 and the data mapping management information 106 are held in the memory. Further, a management terminal 90 having a display screen 92 and an input device 94 is connected to the data mapping management server 72 via a network 46. A keyboard, mouse, or the like is used as the input device 94. The network 46 may be an internal bus, and the display screen 92 and the data mapping management server 72 may be integrally configured.

[0040] The data mapping management program 100 is a program that realizes the functions of the data mapping management server 72. The data storage management information 106 is information necessary for realizing the function of the data mapping management server 72. Specifically, the data storage management information 106 includes mapping information 10a, 10b, 10c, and 10d collected through the management units 104a, 104b, and the management agent program 102 (hereinafter, these are collectively referred to as "mapping information 10". Included) and load information 12a and 12b (hereinafter collectively referred to as "load information 12").

[0041] The data mapping management server 72 receives an interactive instruction from the administrator via the management terminal 90, and executes a process related to the change of the data mapping. At this time, the data mapping management server 72 displays necessary information on the display screen 92. Instructions from the administrator are input via the input device 94.

[0042] Although the data mapping management server 72 is independent of other virtualization mechanisms in FIG. 1, any virtualization mechanism may also serve as the data mapping management server 72.

[0043] FIG. 2 is a diagram showing a hierarchical structure of data mapping according to the embodiment of the present invention. In this figure, the case where two virtualization switches 60 exist between the host 70 and the storage device 50 will be described. Hereinafter, with respect to the two layers, the one closer to the application program 88 is referred to as a higher layer, and the one closer to the disk device 54 is referred to as a lower layer.

[0044] The application program 88 accesses the file 202 provided by the file system 86. The file system 86 secures an area for storing the data of the file 202 in the logical volume 204, and converts the access to the file 202 into the access to the logical volume 204 corresponding to the file 202. Specifically, the address indicating the file 202 included in the received command is converted into the address indicating the logical volume 204, and the command including the address is issued.

[0045] The volume manager 84 converts the access to the logical volume 204 into the access to the area of the virtual volume 206 corresponding to the logical volume 204.

[0046] The virtualization switch 60 converts the access to the virtual volume 206 into the access to the area of the logical disk device 208 or the virtual volume 206 corresponding to the virtual volume 206. The storage device 50 converts the access to the logical disk device 208 into the access to the disk device 54 corresponding to the logical disk device 208.

[0047] Although not shown, certain virtualized data may be shared by a plurality of hosts 70. Data sharing is realized by referencing the same storage area in the lower layer virtual volume 206 or the like to a plurality of upper layer virtualization mechanisms (virtualization switches or the like).

[0048] The virtualization switch 60 is not limited to two stages, and any number of stages may exist. Also, if the file system 86 and volume manager 84 on host 70 do not exist or are not used, if the virtualization switch 60 does not exist, and if the storage device 50 does not retain the logical-physical storage area conversion function, it is a disk device. It is also conceivable that the 54 storage areas are provided to the outside as they are. If the storage device 50 does not have the logical-physical storage area conversion function, it is not necessary to exchange mapping information between the data mapping management server 72 and the storage device 50, so the storage device 50 needs to be connected to the network 46. There is no.

FIG. 3 is a diagram showing a data structure of mapping information 10. Mapping information 10 has entries 302 to 310. In entry 302, information indicating a higher-level virtual structure identifier indicating a virtual structure provided to a higher-level layer by the virtualization mechanism having the mapping table 10 is stored.

[0050] In entry 304, information indicating a block number in the upper structure indicating a storage area in the virtual structure corresponding to the upper virtual structure identifier is stored. In entry 306, information indicating a lower virtualization function identifier indicating a lower virtualization mechanism that provides a storage area to the virtualization mechanism having the present mapping information 10 is stored. Entry 308 contains information indicating a substructure structure identifier that is an identifier of the management structure provided by the virtualization mechanism corresponding to the substructure virtualization function identifier, and entry 310 contains information in the management structure corresponding to the substructure structure identifier. Information indicating the block number in the substructure indicating the area is stored.

[0051] Note that the identifier of the virtualization mechanism used in the lower virtualization function identifier 306 or the like is grasped by the host 70 on which the virtualization mechanism is executed when the virtualization function is the file system 86 or the volume manager 84. What can be used is used.

[0052] Entry 312 whose higher-level virtual structure identifier is "Free" is in a state where the virtualization mechanism having the mapping information can use the area, but has not yet been provided to the higher-level virtualization mechanism. Indicates the storage area of the lower hierarchy. The virtualization mechanism can perform data movement accompanied by copying to the storage area of this lower layer.

FIG. 4 is a diagram showing a data structure of load information 12. The entries 306 to 310 are the same as those described in Mapping Information 10, so the description thereof will be omitted. In entry 314, information indicating a cumulative load information value, which is information on an access load to a storage area provided by a lower-level virtualization mechanism specified in entries 306 to 310, is registered.

[0054] The load information cumulative value is a cumulative value of information related to the load such as the number of accesses to the storage area, the data transfer amount, or the execution time of the data access process from a certain time to a certain time.

[0055] The data mapping management information 106 possessed by the data mapping management server 72 includes the mapping management information 321, the load information aggregation table 322, and the data mapping management setting information 323.

[0056] The data mapping management setting information 323 holds information used for facilitating data mapping management. For example, information for grouping and managing disk devices 54 according to performance, purpose of use, etc. is held.

[0057] FIG. 5 is a diagram showing a data structure of mapping management information 321. The mapping management information 321 includes the hierarchical configuration information 330, the path connection form information 340, and the mapping aggregation information 350.

[0058] The hierarchical configuration information 330 stores information regarding which virtualization mechanism in the lower hierarchy uses the virtual structure provided by the virtualization mechanism. Specifically, the upper-level virtualization function identifier 331, which is the identifier of the upper-level virtualization mechanism, the lower-level virtualization function identifier 332, which is the information for identifying the lower-level virtualization mechanism, and the upper-level layer are the lower-level layers. Includes a set of path ID 334, which is an identifier of the I / O path 40 used for access. In this embodiment, since there is no virtualization mechanism lower than the storage device 50, there is no entry in which the virtualization mechanism indicated by the upper layer virtualization function identifier 331 is the storage device 50.

[0059] In the path connection form information 340, information regarding a path used for data transfer between virtualization mechanisms is registered. Specifically, a set of path ID 334, which is an identifier of the I / O path 40, and connection form information 342, which is information on the transfer speed and connection form of the path, is held. Even if the paths are physically different, only one path is used equivalently. It is also necessary to consider the access routes of file system 86 and volume manager 84 in the same host 70 as paths.

[0060] In the mapping aggregate information 350, information on the data movement function of the virtualization mechanism and the mapping information 10 held by them are collectively stored. Specifically, the virtualization function identifier entry 351 in which the identifier of the virtualization mechanism is registered, and the operation in which information on the minimum unit of data to be moved when the virtualization mechanism has a data movement function accompanied by copying of data is registered. From the mapping change granularity entry 353 and the virtualization mechanism, where information about the smallest unit of mapping change data is registered when the target data movement granularity entry 352, the virtualization mechanism has the function to change the mapping information 10 without copying the data. The virtualization function mapping information entry 354 that holds the collected mapping information 10 is stored.

[0061] Note that the dynamic data movement particle size entry 352 and the mapping change particle size entry 353 hold invalid values when the virtualization mechanism corresponding to the entry does not have these functions.

[0062] FIG. 6 is a diagram showing a data structure of the load information aggregation table 322 included in the data mapping management information 106. In the load information aggregation table 322, the virtualization function identifier entry 351 in which the identifier of the virtualization mechanism that collects the load information 12 is registered, and the lower level in which the information indicating the storage area of the lower hierarchy in which the load information 12 is collected is registered. Virtualization function identifier entry 306, lower management structure identifier entry 308 and lower structure block number entry 310, old load information cumulative value entry 364 that holds the load information cumulative value 314 of the corresponding area when the load information 12 was collected last time. Includes the load information aggregated value entry 365 in which information about the load in the corresponding area within a certain period is registered, and the previous load information cumulative value acquisition time 366, which is the time when the previous load information 12 was collected.

[0063] As described above, the storage device 50 and the virtualization switch 60 have a function of providing the load information 12 to the external device. Utilizing this, the data mapping management server 72 acquires the load information 12 from the virtualization mechanism at regular time intervals through the network 46, and the accumulated load information 12 is accumulated as the load information cumulative value 314 and the old load information cumulative value. Based on 364, the previous load information cumulative value acquisition time 366, and the current time information, the load information from the time 366 to the current time is calculated and stored in the corresponding location of the load information aggregate value 365.

After that, the data mapping management server 72 stores the acquired load information cumulative value 314 in the corresponding old load information cumulative value entry 364, and further stores the current time used in the calculation in the previous load information cumulative value acquisition time entry. Store in 366. The content of the load information stored in the load information aggregated value entry 365 differs depending on the content of the load information cumulative value entry 314. For example, the number of I / Os per unit time, the amount of data transfer, the cumulative access time, etc. can be considered. ..

[0065] Note that it is not necessary for all the storage devices 50 and the virtualization switch 60 to collect the load information 12. In this case, only the information obtained from the collected load information 12 is registered in the load information aggregation table 322. If neither the storage device 50 nor the virtualization switch 60 collects the load information 12, the load information aggregation table 322 does not have to exist.

[0066] Hereinafter, the data mapping change process in the present invention will be described. The first is the data movement process that involves copying the data, the second is the mapping modification process that does not involve copying the data, and the third is the garbage that modifies the data mapping to free up areas that do not hold valid data. Collection (GC) processing. The data mapping management server 72 shall issue a series of instructions regarding the data mapping change process. Further, even if the configuration of the virtualization mechanism and the number of layers are different from those described below, the same processing as in the following description is performed.

[0067] FIG. 7 is a diagram illustrating an outline of data movement in a storage area with a copy of the data. This figure shows an example in which the mapping between the virtual volume 206a constructed by the virtualization switch 60a and the lower hierarchy is changed by data movement accompanied by copying.

[0068] Before the mapping is changed, the data of the virtual volume 206a is assigned to the virtual volume 206b provided by the virtualization switch 60b. Further, the virtual volume 206b is further assigned to the logical disk device 208b configured by the disk device 54b in the storage device 50b.

[0069] In the present embodiment, the mapping of the virtual volume 206a is changed to the virtual volume 206c in the virtualization switch 60c by moving the data stored in the disk device 54b to the disk device 54c.

[0070] When the mapping is changed, if the virtual volume 206c does not exist in the virtualization switch 60c, the virtual volume 206c is created in the virtualization switch 60c. First, the data mapping management server 72 issues an instruction to the storage device 50c to create the logical disk device 208c from the available storage area among the storage areas of the disk device 54c of the storage device 50c. After that, the virtualization switch 60c is instructed to recognize the created logical disk device 208c.

Subsequently, the data mapping management server 72 issues an instruction to the virtualization switch 60c to create the virtual volume 206c using the logical disk device 208c. After that, the virtualization switch 60a is instructed to recognize the virtual volume 206c. Then, the virtualization switch 60a is instructed to perform data movement accompanied by copying and change the storage area of the virtual volume 206a allocated in the virtual volume 206b to the virtual volume 206c.

After that, the data mapping management server 72 performs GC processing on the virtual volume 206b when another area of the virtual volume 206b is used by another virtualization mechanism. If the other area of the virtual volume 206b is not used by another virtualization mechanism, the virtualization switch 60b is instructed to delete the virtual volume 206b.

[0073] Then, the data mapping management server 72 leaves the entry corresponding to the deleted virtual volume 206b in the mapping information 10b of the virtualization switch 60b in the Free state. Alternatively, the virtualization switch 60b is instructed to stop using the logical disk device 208b, and then the storage device 50b is instructed to delete the logical disk device 208b.

[0074] When moving data with copying, if the data to be moved is shared by a plurality of virtualization mechanisms, maintaining data consistency by a plurality of access paths during the data movement process becomes a problem. .. In order to avoid this problem, at the time of data movement, mapping change processing without copying is performed so that the virtualization layer in which data sharing is performed is higher than the layer in which data movement is performed. Then, after the movement of the data is completed, if necessary, the mapping change process without copying to restore the data sharing hierarchy is performed.

[0075] Further, when allocating the storage area of the logical disk device 208c in FIG. 7, the disk device 54 corresponding to the storage area marked as Free in the mapping information 10 managed by another virtualization mechanism. The storage area inside is available. In this case, the data mapping management server 72 instructs the corresponding virtualization mechanism and the storage device 50 to open the corresponding storage area and allocate the storage area corresponding to the storage device 50 to the logical disk device 208c. .. In addition, GC processing may be involved in opening this area.

[0076] FIG. 8 is a diagram illustrating an outline of a mapping change without copying the data. In this figure, the configuration of the virtual volume 206d provided by the virtualization switch 60d is changed by the mapping change without copying.

[0077] Before the mapping change, the virtual volume 206d is mapped to the virtual volume 206e provided by the virtualization switch 60e. Further, the virtual volume 206e is mapped to the logical disk device 208d composed of the disk device 54d in the storage device 50d. This state is changed to the logical volume 206f in the virtualization switch 60f by the mapping change without copying.

[0078] First, the data mapping management server 72 searches for the mapping management information 321 and finds the area of the logical disk device 208d corresponding to the virtual volume 206d in the virtualization switch 60e whose mapping is changed. The data mapping management server 72 instructs the virtualization switch 60f to create a virtual volume 206f consisting of the obtained area of the logical disk device 208d. After that, the data mapping management server 72 suspends the processing of I / O execution for the virtual volume 206d on the virtualization switch 60d, and virtualizes the part assigned to the virtual volume 206e of the virtual volume 206d in the virtual switch 60d. Change to assign to volume 206f and instruct virtual volume 206d to resume I / O execution processing.

After that, when the virtual volume 206e is not used at all by other virtualization mechanisms, the data mapping management server 72 instructs the virtualization switch 60e to delete the entry corresponding to the virtual volume 206e. .. If there is a virtualization mechanism that still uses a part of the virtual volume 206e, the data mapping management server 72 performs GC processing related to the virtual volume 206e.

[0080] FIG. 9 is a diagram showing an outline of the GC process. The left side of this figure shows that the virtualization switch 60g configures the virtual volume 206g by using the area 251b in the logical disk device 208g provided by the storage device 50g. Further, the area 251a in the same logical disk device 208g is allocated to the disk device 54h, and the area 251b is allocated to the disk device 54g.

[0081] Here, it is assumed that the area 251a is not used by another virtualization mechanism. In this case, since the logical disk device 208g is provided to the virtualization switch 60g, when the disk device 54h corresponding to the area 251a is used by another virtualization mechanism, the area 251a is accessed from the virtualization switch 60g. The possibility cannot be denied. The reverse is also possible.

Therefore, a GC process is performed to open the area 251a with respect to the logical disk device 208g. First, the data mapping management server 72 obtains the area 251b.

[0083] When the data mapping change function without copying is available, the data mapping management server 72 instructs the storage device 50g to create the logical disk device 208i consisting of the area 251b. After that, the virtualization switch 60g is instructed to change the data mapping without copying so as to configure the virtual volume 206g by using the logical disk device 208i.

[0084] When the data mapping change function without copying is not available, the data mapping management server 72 first instructs the storage device 50g to create a logical disk device 208i consisting of an area of the same size as the area 251b. put out. After that, the virtualization switch 60g is instructed to move data with a copy so as to change the storage area of the virtual volume 206g to the logical disk device 208i. Then, a data movement instruction accompanied by copying is issued to the storage device 50g to change the storage area of the logical disk device 208i to the storage area of the original area 251b. It is not always necessary to change the mapping in the last storage device 50g.

[0085] When the storage area to be subjected to the GC processing is shared by a plurality of virtualization mechanisms, the shared data may cause a consistency problem due to data movement accompanied by copying. At that time, if the same solution as data movement involving copying cannot be obtained, such GC processing should not be performed.

[0086] FIG. 10 is a diagram showing a procedure of mapping update processing. In order to simplify the explanation, the process related to the error is omitted in this procedure. If an error occurs during the following processing, the processing to return to the original configuration is performed if possible. Further, this processing is performed by the data mapping management server 72 controlling each virtualization mechanism.

[0087] First, in order to avoid problems due to data sharing, a hierarchy in which data sharing is performed as necessary according to the instructions of the data mapping management server 72.<u style="single">Shared area data</u>Is changed to a layer higher than the layer in which the mapping change process is performed. This is a mapping without the copy mentioned above<u style="single">Change processing</u>Is realized by (step 1202).

Next, when the data movement accompanied by the copy shown in FIG. 7 is performed, the management structure to be the movement destination is assigned if necessary. As described above, the area marked Free in the mapping information 10 may be opened and used as the destination. In this case, GC processing may be involved. (Step 1203).

After that, based on the instruction of the data mapping management server 72, the data movement process with copying and the mapping update process without copying are executed (step 1204). Next, GC processing is performed as needed. At this time, when processing using the data movement function is performed, data sharing is performed in a layer higher than the layer in which the data movement is performed, and after the data movement is completed, in the original layer as necessary. The process of returning to the data sharing of is performed (step 1205). After that, if necessary, the mapping update function that does not involve copying performs a process of moving the hierarchy in which data sharing is performed to a desired location (step 1206).

[0090] Hereinafter, the screen that the data mapping management server 72 shows to the administrator when the data mapping change process is executed will be described. This screen is displayed on the output screen 92 held by the management terminal 90. Then, the data mapping management server 72 receives an instruction from the administrator based on this screen from the input device 94.

[0091] Various mapping information is presented on the screen described below. At that time, the necessary mapping information is obtained by referring to the data mapping management information 106 (particularly the mapping management information 321) and matching the corresponding parts. Further, when displaying other information such as the setting function, it is requested by referring to the data mapping management information 106.

FIG. 11 is a diagram showing a management main screen 500 of a computer system in which the data mapping management server 72 manages data mapping, which is displayed on the output screen 92. On the screen 92, a system configuration display area 502 on which the computer system is displayed, a load information display period display area 504, a button 506 operated by a pointer or the like when opening various screens, and a close to erase the management main screen 500. Button 508 is displayed.

[0093] The management main screen 500 is the screen that is first displayed when the administrator performs the data mapping change process. When the administrator finishes the process, he / she selects the close button 508 to delete the management main screen 500 from the screen.

[0094] In the system configuration screen display area 502, the configuration of the computer system in which the data mapping management server 72 manages the data mapping is illustrated. When the administrator selects a picture showing the storage device 50, virtualization switch 60, and host 70 drawn in this area with the pointer on the screen (for example, double-clicking with the mouse), for example, the load corresponding to the selected device Information about the device, such as the information in the information aggregation table 322, is displayed on the display screen 92.

[0095] When the storage device 50 or the virtualization switch 60 collects the load information 12, the data mapping management server 72 sets the virtual volume 206 and the logical disk during the period displayed in the load information display period display area 504. The load information related to the device 208, the disk device 54, and the I / O path 40 is obtained by summarizing the corresponding data in the load information aggregation table 322 with reference to the mapping management information 321.

[0096] When there is a device whose load information exceeds a predetermined threshold value, the data mapping management server 72 is high for the administrator by using means such as changing the drawing color of the picture corresponding to the device. Present the part that is the load.

[0097] When the administrator selects the button 506a with a pointer or the like, the file browsing screen 510 described later is displayed on the display screen 92. When the button 506b is selected, the virtual storage area browsing screen 520 described later is displayed on the display screen 92. When the button 506c is selected, the physical storage device browse screen 540 described later is displayed on the display screen 92. When the button 506d is selected, the display screen 92 displays a screen for grouping and other settings for facilitating the management of the load information display period 504 and the disk device 54.

[0098] FIG. 12 is a diagram showing a file system browse screen 510, which is a screen for presenting a directory structure managed by a file system 78 existing in a computer system to an administrator. The file system browse screen 510 displays a file system browse area 512, which is an area for displaying the directory structure, and a close button 508, which is selected when the file system browse screen 510 is deleted from the screen.

[0099] In the file system browse area 512, the directory structure used for each host 70 is displayed in a tree structure. In addition, a scroll bar selected by the pointer when scrolling the display area when the data cannot be displayed is added to the area. When the administrator selects (double-clicks) entry 516 displayed in the file system browse area 512 with the pointer, detailed information on the corresponding file 202 is displayed on screen 92.

The mark 514 indicates that the corresponding entry 516 meets the conditions associated with a particular virtualization mechanism or management structure. This is commonly used with the virtual storage area browse screen 520 and the physical storage device browse screen 540, which will be described later. Information about the mark 514 is stored in the data mapping management setting information 323. The data mapping management server 72 seeks a management structure that satisfies the condition while referring to the mapping management information 321 and displays the mark 514. If the conditions corresponding to the marks are different from each other, the display color and pattern of the mark 514 should be different. Due to the nature of mark 514, multiple mark 514 may appear in the same entry.

[0101] The setting of the mark 514, which means that a specific file, for example, the above-mentioned file 202 is included, is displayed on the screen by the administrator by selecting (right-clicking) the entry 516 corresponding to the file 202 with the pointer. Is displayed and the "mark setting" process in it is selected. Similarly, the setting of the mark 514 indicating access by the host 70 can be similarly performed in the file system browse area 512.

[0102] FIG. 13 is a diagram showing a virtual storage area browsing screen 520, which is a screen displaying the logical volume 202, the virtual volume 204, and the logical disk device 206 existing in the computer system.

[0103] On the virtual storage area browse screen 520, a close button 508 selected by the administrator when the logical storage area browse area 522, the logical storage area information display area 526, and the virtual storage area browse screen 520 are deleted from the screen is displayed. Will be done. A plurality of virtual storage area browsing screens 520 can be displayed on the display screen 92 at the same time.

[0104] The logical storage area browse area 522 is provided with a scroll bar selected by a pointer to scroll the display area when data cannot be displayed in the area. In the logical storage area browse area 522, the logical volume 202, the virtual volume 204, and the logical disk device 206 existing in the system are grouped and displayed for each virtualization mechanism that provides them. When the administrator selects (clicks) the displayed entry 524 with the pointer, the information of the corresponding logical volume 202, virtual volume 204, and logical disk device 206 is displayed in the logical storage area information display area 526. The above-mentioned mark 514 is also used for entry 524. The setting method of the mark 514 is also the same as described above.

[0105] The information of the logical volume 202, the virtual volume 204, and the logical disk device 206 selected from the display is displayed in the logical storage area information display area 526. Specifically, the name of the virtualization mechanism that provides the selected logical volume, etc., the name of the selected logical volume, etc. and its storage capacity, the related management structure display selection area 528, the related management structure name list display area 530, The logical block mapping display area 534 and the detailed information display button 536 are displayed.

[0106] The administrator selects the type of the related management structure to be displayed in the related management structure name list display unit 530 and the logical block address mapping display area 534 from the display of the related management structure display selection area 528. The related management structure refers to a management structure that directly or indirectly uses the storage area provided by a certain management structure, and a management structure that directly or indirectly provides the storage area to a certain management structure.

[0107] For example, when the administrator selects (clicks) this area 528, the types of related management structures such as "direct component management structure", "top-level logical volume", and "physical storage area" are selected. A list is displayed. The administrator specifies the management structure to be used for display by further selecting (clicking) the entry in the entry.

[0108] The related management structure name list display unit 530 displays a list having the name of the type of management structure selected by the related management structure display selection unit 528 as an element. The related management structure name list display unit 530 is provided with a scroll bar for instructing scrolling of the display area when the data cannot be displayed in the area. The mark 514 is also added to the entry displayed on the related management structure name list display unit 530 depending on the content of the data held in the storage area corresponding to the entry.

[0109] In the logical block address mapping display unit 534, the storage area in the management structure in which the information is displayed corresponds to the data of any part of the related management structure of the type selected by the related management structure display selection unit 528. Whether to do so is displayed by dividing it into the related management structure corresponding area 538. The mark 514 is also added to the related management structure corresponding area 538 depending on the content of the data to be held. Further, the mark 514 corresponding to the mark 514 can be set by the same method as described above. Further, when the corresponding area information can be directly displayed in characters, it is displayed. At this time, the area corresponding to the management structure that can be used as the data copy destination is also displayed to that effect.

[0110] When the detailed information display button 536 is selected on the screen, the logical volume or the like currently displayed on the logical storage area information display unit 526 is assigned to the management structure such as the synchronous data copy function. A screen for displaying detailed information such as the function, the load information in the load information aggregation table 322 related to the function, the high-resolution information of the same information as the logical block address mapping display unit 534 is displayed.

[0111] FIG. 14 is a diagram showing a physical storage device browse screen 540, which is a screen displayed on the output screen 92 and displays a disk device 54 existing in the computer system. On the physical storage device browse screen 540, the physical storage device browse unit 542, the physical storage device information display unit 560, and the close button 508 used when erasing the physical storage device browse screen 540 are displayed. A plurality of physical storage device browse screens 540 can be displayed on the output screen 92 at the same time.

[0112] The physical storage device browsing unit 542 is a part that displays the disk device 54 existing in the system, in which the group list display unit 544 and the physical storage device list display unit 548 are displayed, and further, the filter setting unit 552. And has a mark information display button 554.

[0113] The group list display unit 544 is provided with a scroll bar for instructing scrolling of the display area when data cannot be displayed in the display area. The group list display unit 544 displays the entire system, the host 70, the virtualization switch 60, the storage device 50, the logical volume 202, the virtual volume 204, the logical disk device 206, the group set for management, and the like. .. At this time, the logical volume 202, the virtual volume 204, and the logical disk device 206 are grouped by the virtualization mechanism that provides them, and each component of the group is displayed as a subentry thereof. When the entry 546 displayed in the group list display unit 544 is selected (clicked) on the screen, the disk device 54 included in the group corresponding to the entry 546 is filtered by the setting displayed in the filter setting unit 552. The list of the disk devices 54 after the processing is performed is displayed on the physical storage device list display unit 548. Mark 514 is added to entry 546 if it holds the corresponding data.

[0114] The physical storage device list display unit 548 displays a list of the names of the disk devices 54 obtained by the above method. Also here, a scroll bar is added to instruct scrolling of the display area when the data cannot be displayed in that area. When the entry 550 displayed on the physical storage device list display unit 548 is selected, the information of the corresponding disk device 54 is displayed on the physical storage device information display unit 560. The mark 514 is also added to the entry 550 according to the data to be held.

[0115] The filter setting unit 552 specifies the filter processing to be performed when displaying the list of the disk devices 54 on the physical storage device list display unit 548. When the check box of the Free entry is checked, when the filtering process is executed, the area corresponding to the entry 312 named Free is displayed in the mapping information 10 in one of the virtualization mechanisms. The process of selecting only the including disk device 54 is performed. If the check boxes for the Mark1, Mark2, Mark3, and Mark4 entries are checked, only disk devices 54 that contain data that meets the conditions identified by the mark 514 drawn beside them are selected as filtering. The process is carried out. When a plurality of entries are checked, a filter process is performed to select only the disk devices 54 that satisfy all the checked entries.

[0116] When the mark information display button 554 is selected, the conditions for adding the mark information 514 are displayed on the screen. A shortcut operation that can be displayed from other screens may be assigned to this screen display operation.

[0117] The setting of the mark 514, which means that the data related to the disk device 54 is included, is performed by the same method as described above. Similarly, the group list display unit 544 also sets the mark 514 indicating that the data related to the logical volume 202, the virtual volume 204, the logical disk device 206, the host 70, the virtualization switch 60, or the storage device 50 is included. Can be done.

[0118] The information of the selected disk device 54 is displayed on the physical storage device information display unit 560. Specifically, the name of the storage device 50 that holds the selected disk device 54, the name of the selected disk device 54 and its storage capacity, the related management structure display selection unit 528, the related management structure name list display unit 530, The logical block mapping display unit 534 and the detailed information display button 562 are displayed.

[0119] When the detailed information display button 562 is selected, the disk device 54 currently displayed on the logical storage area information display unit 526 is allocated to the area of the management structure such as the synchronous data copy function. Detailed information such as the function, the load information in the load information aggregation table 322 related to the function, and the high-resolution information of the same information as the logical block address mapping display unit 534 is displayed.

[0120] The data mapping change instruction is given as follows using the virtual storage area browse screen 520 and the physical storage device browse screen 540. The data to be moved is specified by the related management structure corresponding area 538 displayed on the logical block address mapping display unit 534. Further, the data mapping change destination is instructed by dragging and dropping the display on the screen corresponding to the data whose mapping is changed.

[0121] At this time, by selecting a plurality of related management structure corresponding areas 538 while pressing an appropriate key such as a keyboard, a function for simultaneously selecting a plurality of related management structure corresponding areas 538 is provided, and they are combined. At the same time, it is possible to specify the data mapping change destination.

[0122] The valid parts as the mapping change destination are as follows. The first is another related management structure corresponding area displayed on the entry 550 in the physical storage device list display unit 548 on the physical storage device browse screen 540 or the logical block mapping display unit 534 in the physical storage device information display unit 560. It is 538. In this case, the data is moved to the area in the disk device 54 corresponding to the specified entry 550 or the disk device 54 corresponding to the related management structure corresponding area 538.

The second is another related management structure displayed in the entry 524 in the logical storage area browse unit 522 and the logical block mapping display unit 534 in the logical storage area information display unit 526 on the virtual storage area browse screen 520. It is the entry 546 in the group list display unit 544 in the corresponding area 538 or the physical storage device browse screen 540. In this case, the data is specified to be moved through the specified virtualization mechanism, management structure, or corresponding area within the management structure in the mapping hierarchy described conceptually in FIG. ..

[0124] In relation to the mapping change instruction, the menu displayed when the related management structure corresponding area 538 is selected with the right mouse button includes the following items. The first is the "division" process that divides the related management structure corresponding area 538. This is possible only if the virtualization mechanism that provides the selected related management structure corresponding area 538 supports the division of the dynamic management structure. When this is selected, a screen is displayed to specify how to divide the selected related management structure corresponding area 538. When instructed on the screen, the virtualization mechanism is instructed to actually perform the division process. After the instruction is successful, the related management structure corresponding area 538 is divided and displayed. By using this division process, it is possible to change the mapping of only a part of an arbitrary area.

[0125] The second item is for issuing a data mapping change instruction of the selected related management structure corresponding area 538 on the screen displaying the mapping related thereto. When this is selected, the following data mapping change display screen 580 is displayed.

[0126] FIG. 15 displays the data mapping change processing proposal created by the data mapping management server 72 displayed on the output screen 92, or displays the data mapping related to the related management structure corresponding area 538. The data mapping change display screen 580 used when issuing a mapping change instruction on the screen is shown.

On the data mapping change display screen 580, select when deleting the data mapping change processing procedure proposal display unit 582, the data mapping change related management structure browsing unit 584, the data mapping change editing unit 586, and the data mapping change display screen 580. The close button 508 to be displayed, the button 596a selected when instructing the execution of the data mapping change process, and the button 596b selected when instructing to stop the process are displayed.

[0128] In the data mapping change processing procedure proposal display unit 582, the processing procedure proposal for realizing the data mapping change displayed in the data mapping change editing unit 586 is displayed according to the execution order. Further, the display unit 582 is provided with a scroll bar for instructing scrolling of the area when the data cannot be displayed in the display area. If the data mapping change processing procedure cannot be created, a message to that effect is displayed. The data mapping change processing procedure draft display unit 582 displays the processing procedure for realizing the specified data mapping by the mapping change function each time the data mapping is modified.

[0129] The displayed processing procedure is obtained as follows. First, the data mapping management server 72 confirms whether the specified mapping can be realized by the processing procedure of only the mapping change without copying in each virtualization mechanism. Specifically, for parts where the mapping is different before and after the change, is it possible to create a management structure corresponding to the mapping after the change, or is it possible to change the mapping without copying in the virtualization mechanism where the mapping branches due to the change? It is confirmed whether it is feasible.

[0130] When the storage area on the disk device 54 is changed after the mapping is changed, it is confirmed whether or not it can be realized by the processing procedure including the data movement accompanied by copying. Specifically, for the parts where the mapping is different before and after the change, can the management structure corresponding to the mapping after the change be created so as to correspond to the storage area on the disk device 54 allocated as the copy destination? It is confirmed whether it is possible to carry out data movement with copying by the virtualization mechanism where the mapping branches. The mapping change processing procedure created at these times is as described above. It also includes a mapping change processing procedure for GC processing.

[0131] What needs attention here is the process related to data sharing when data transfer accompanied by copying is performed. As described above, it is necessary to share data in a hierarchy higher than the hierarchy in which data is moved with copying, and if data sharing in the higher hierarchy is not possible, the specified data mapping Is judged to be unrealizable.

[0132] Further, the particle size when actually changing the mapping with reference to the mapping management information 321 is also taken into consideration.

[0133] Further, the mapping change target may be separated into a plurality of management structures in a different hierarchy. In this case, the above confirmation conditions are satisfied for all related management structures.

[0134] The data mapping change-related management structure browse unit 584 is provided with a scroll bar for instructing scrolling of the display area when the data cannot be displayed in the display area. In the browse unit 584, the related area of the management structure related to the data mapping change displayed in the data mapping change editing unit 586 is displayed in the mapping tree structure from the logical volume 204 existing in the host 70. When the corresponding entry displayed is selected, a screen is displayed showing detailed information about the relevant area.

[0135] The data mapping change editing unit 586 is provided with a scroll bar for instructing scrolling of the display area when the data cannot be displayed in the display area. Then, in the editorial department 586, the data mapping before and after the change related to the related management structure corresponding area 538 that changes the data mapping is displayed with the icon 588a indicating the related area of the management structure after the mapping change and the management structure before the mapping change. It is displayed using the icon 588b, which indicates the relevant area.

[0136] The administrator instructs to change the mapping to a mapping via another virtualization mechanism by dragging and dropping the icon 588a to an appropriate place. The change in the data arrangement in the disk device 54 is displayed in the entry 550 in the physical storage device list display unit 548 in the physical storage device browse screen 540 or the logical block mapping display unit 534 in the physical storage device information display unit 560. Related management structure This is done by dragging and dropping the icon 588a into the corresponding area 538. Then, when the icons 588a and 588b are selected (double-clicked), the physical storage device browse screen 540 that displays the information of the disk device 54 corresponding to the selected icons 588a and 588b is displayed on the display screen 92.

[0137] FIG. 16 is a flowchart showing a procedure of data mapping change processing performed interactively with the data mapping management server 72 by the administrator via the management terminal 90. The process shown in FIG. 16 is interrupted when the close button 508 in the management main screen 500 is selected, except when the process shown in step 1012 is executed.

[0138] The administrator starts the data mapping change process (step 1001). The data mapping management server 72 collects mapping information 10 as work information for performing this process, and updates the mapping management information 321 so as to retain the latest information (step 1002).

[0139] The data mapping management server 72 refers to the mapping management information 321 to determine which storage area of which disk device 54 is an area that can be used for data movement change accompanied by copying, and creates the information. To do. This is because the data mapping management server 72 refers to the virtualization function mapping information 354, grasps the data copyable area in each virtualization function, and confirms which disk device 54 the area corresponds to. Do (step 1003).

[0140] The data mapping management server 72 presents information on the system configuration and load to the administrator via the management terminal 90, and receives a data mapping change instruction. This process is executed using the management main screen 500, the file browse screen 510, the virtual storage area browse screen 520, and the physical storage device browse screen 540, and the details are as described when these screens are explained. In addition, there is a case where the data mapping change display screen 580 is opened from the related management structure corresponding area 538, but in that case, it is assumed that the process of step 1005 is performed assuming that the data mapping without change is specified (). Step 1004).

[0141] The data mapping management server 72 executes a process of creating a process procedure draft of a data mapping change that satisfies the data mapping specified in step 1004. If no data mapping change is specified, the process is considered successful as no process procedure (step 1005). The data mapping management server 72 confirms whether the draft processing procedure for the data mapping change succeeded or failed (step 1006). If it fails, the administrator is notified that the process has failed and the cause thereof, and then the process returns to step 1004 (step 1021).

If successful, the data mapping management server 72 presents the proposed data mapping change processing procedure created in step 1005 to the administrator using the data mapping change display screen 580 (step 1007). ..

After that, the data mapping management server 72 accepts permission / correction / cancellation of the created data mapping change processing procedure using the data mapping change display screen 580. The details are as explained on the data mapping change display screen 580 (step 1008).

[0144] It is confirmed that the instruction received in step 1008 is canceled, that is, whether the button 596b is selected on the data mapping change display screen 580. If canceled, the process returns to step 1004 (step 1009).

If the instruction is not cancelled, the data mapping management server 72 checks whether the data mapping can be changed depending on whether or not a processing procedure for realizing the specified data mapping can be created. The judgment is made when requesting the content to be displayed on the data mapping change processing procedure draft display unit 582 (step 1010). If the data mapping cannot be changed, the data mapping management server 72 is corrected. Notify the administrator that the data mapping of the above cannot be realized, and then return to the process of step 1008 (step 1022).

[0146] If the data mapping can be changed, does the data mapping management server 72 really carry out the data mapping change according to the data mapping change processing procedure currently displayed on the data mapping change display screen 580? Make a final confirmation (step 1011). If the change is canceled, the data mapping management server 72 returns to the process of step 1008.

[0147] When the execution of the data mapping change is selected, the data mapping management server 72 executes the data mapping change processing according to the created processing procedure (step 1012). After completion, the data mapping management server 72 notifies the administrator via the management terminal 90 that the data mapping change process has been completed due to success or failure (step 1013), and configures the mapping management information 321 after the change process is completed. Update to the one and go back to step 1003 (step 1014).

[0148] FIG. 17 is a flowchart showing a processing procedure for creating a data mapping change processing procedure draft in step 1005 of FIG. The data mapping management server 72 starts processing. At this time, it is assumed that information on the related management structure corresponding area 538 that is the target of the data mapping change and the change destination is given (step 1101).

[0149] After that, the data mapping management server 72 confirms whether or not the instructed data mapping change target can be assigned to the change destination from the following viewpoints. If the mapping change destination is the storage area in the disk device 54, the specified area does not hold valid data (marked as "Free" in the mapping information 10 of a virtualization mechanism). (It is the storage area in the disk device 54 corresponding to the stored storage area), and check whether the specified area has enough capacity to hold the data mapping change target.

[0150] When the one specified as the mapping change destination is the disk device 54, there is an area in which the valid data is not held, and the area holds a capacity sufficient to hold the data mapping change target. Check if you are doing it.

[0151] When another thing is specified as the mapping change destination, it is considered that it can be assigned to the change destination.

[0152] In the above confirmation process in step 1102, the particle size when actually changing the mapping is taken into consideration (step 1102).

[0153] When the instructed data mapping change target cannot be assigned to the change destination, the data mapping management server 72 reports to the administrator the failure to create the data mapping change processing procedure, and ends the processing ( Step 1131) If the indicated data mapping change target can be assigned to the change destination, the data mapping management server 72 is in the hierarchy of other virtualization functions related to the related management structure corresponding area 538 that is the data mapping change target. Understand the relevant areas of management structure and their mapping. This process is performed by referring to the mapping management information 321. This mapping may be separated into multiple management structures in a certain hierarchy, but from this point onward, all of them shall be grasped (step 1103).

After that, the data mapping management server 72 confirms whether the same area is shared by a plurality of hosts in the hierarchy lower than the hierarchy to which the data mapping is changed for the mapping of the related management structure grasped in step 1103 (step). 1104).<u style="single">The same area is from multiple hosts in the hierarchy below the hierarchy to which the data mapping is changed.</u>If not shared, the data mapping management server 72<u style="single">In the relevant management structure identified in step 1103</u>If there is no shared hierarchy, all hierarchies are included, and if there is at least one shared hierarchy, the hierarchy below the lowest shared hierarchy is included as an entry. Create a modifiable hierarchical list of (step 1121).

[0155] If determined to be shared in step 1104, the data mapping management server 72 examines the data mapping from multiple hosts sharing the data and changes the mapping without copying.<u style="single">The data in the shared area</u>Check if data sharing is possible in the upper hierarchy, and if possible, check whether the data sharing hierarchy is higher than the hierarchy to be changed by this mapping change. This process is performed with reference to the mapping management information 321 (step 1105). If it cannot be done, the process is completed as a creation failure (step 1131).

[0156] If it is determined in step 1105 that the data mapping management server 72 comes to the upper level, the data mapping management server 72 is set to the shared hierarchy.<u style="single">Data in the shared area of is in the hierarchy that can be shared at the highest level</u>Ask for a mapping modification method that does not involve a copy to modify and set it as the first part of the processing procedure. After that, a mapping changeable hierarchy list is created by the hierarchy existing below the upper hierarchy (step 1106).

After that, the data mapping management server 72 changes the mapping in any of the hierarchies included in the changeable hierarchy list obtained so far, and creates all possible mapping change proposals that pass through the mapping change destination. Regarding them, check whether the processing procedure can be created by the processing procedure creation method described in the explanation of the data mapping change processing procedure proposal display unit 582. If the area specified in the disk device 54 or the disk device 54 as the mapping change destination is a disk device 54 different from the change source, data movement accompanied by copying is indispensable. Otherwise, a pattern with only mapping changes without copying and a pattern with copying to an area in disk device 54 that is wider than the data mapping change target and does not hold valid data. There are two possible types. At this time, all possible data mapping proposals and processing procedures for their realization are obtained (step 1107).

[0158] The data mapping management server 72 for which the processing procedure has been obtained confirms in step 1107 whether or not even one new data mapping and the proposed realization processing procedure have been successfully created (step 1108). If none of them succeed, the process is completed as a creation failure (step 1311).

[0159] When even one processing procedure proposal is successfully created, the data mapping management server 72 confirms the number of created processing procedure proposals, and when a plurality of processing procedure proposals are created, it is the most appropriate among them. Perform the process of selecting the appropriate one. The following are adopted as the selection method.

[0160] First, if the load information aggregation table 322 does not exist, or if sufficient information does not exist there, the following items are examined and selected in the following order of priority.

[0161] The data movement process involving copying is performed the least number of times. (2) Data movement processing that involves copying is performed only within the storage device 50. (3) Below, the one with the smallest number of layers from the storage device 50 of the layer that performs data movement processing involving copying.

[0162] If there are a plurality of items having the same rank, appropriately select them. If sufficient information exists in the load information summary table 322, the load on the disk device 54 to be changed after changing to the new data mapping and the I / O path 40 in that mapping is displayed in the mapping summary information 350. And the corresponding data are summarized while referring to the hierarchical configuration information 330.

[0163] At this time, if the unit of the data for which the load information is collected is larger than the data to be changed in the data mapping, "distribute in proportion to the size of the data area" or consider the worst case. Then, corrections such as "assuming that the same load as measured in the data collection unit is independently applied in all areas belonging to the data collection unit" are corrected.

[0164] After that, the data mapping management server 72 excludes the above-mentioned aggregated values in the new data mapping that exceed a predetermined threshold value at a certain part, and the same criteria as when the load information aggregated table 322 does not exist thereafter. Select with. If all are excluded by the threshold, the one with the smallest difference from the threshold is selected (step 1109).

[0165] After that, the data mapping management server 72 completes the process as the data mapping change processing procedure draft creation process is successful. It is also possible to perform step 1107 and step 1109 at the same time (step 1110).

[0166] In the data mapping change processing procedure draft creation process from step 1101, the confirmation of the data sharing hierarchy and the higher hierarchy thereof are confirmed, but the processing of restoring the hierarchy is not included in the processing procedure. This is because it is necessary to leave it to the administrator to determine at which layer data sharing is appropriate when data is moved with copying. Therefore, after this process is completed, the proposed process procedure shall be presented to the administrator, at which time the administrator shall manually modify and specify the data mapping.

[0167] According to the present invention, there is a function of providing a virtual storage area in a plurality of devices, and a desired data mapping is instructed in a computer system in which a function of changing data mapping is provided in part or all of them. Then, the function of creating a detailed processing procedure for realizing the data mapping is provided. At this time, the designated data may be shared and used by a plurality of computers. In addition, the user can intuitively instruct the desired data mapping using the GUI. With this function, the data arrangement desired by the administrator can be easily realized, and the management cost of data mapping is reduced.

[0168] Further, when creating a data mapping change processing procedure, the data storage position change function provided by the plurality of devices grasps all the areas that can be used as the mapping change destination, and uses them as necessary. It is also used for mapping change processing in other devices. As a result, the utilization efficiency of the storage area of the storage device in the computer system is improved.

[0169] Further, when the load information is available, a processing procedure for realizing the data mapping in consideration of the load information is created, so that better data mapping can be obtained.

[Effect of the Invention] According to the present invention, the data arrangement desired by the administrator can be easily realized, so that the management cost of data mapping can be reduced.

[0171] Further, according to the present invention, the utilization efficiency of the storage area of the storage device in the computer system is improved.

[0172] Further, according to the present invention, better data mapping in a computer system can be obtained.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram showing a configuration of a computer system according to an embodiment of the present invention.

FIG. 2 is a diagram showing a concept of a hierarchical structure of data mapping according to an embodiment of the present invention.

FIG. 3 is a diagram showing a data structure of mapping information 10.

FIG. 4 is a diagram showing a data structure of load information 12.

FIG. 5 is a diagram showing a data structure of mapping management information 321.

FIG. 6 is a diagram showing a data structure of a load information aggregation table 322.

FIG. 7 is a schematic explanatory view of a data movement process involving copying of data.

FIG. 8 is a schematic explanatory diagram of a mapping change process that does not involve copying data.

FIG. 9 is a schematic explanatory diagram of a GC process in which data mapping is changed in order to open an area that does not hold valid data.

FIG. 10 is a diagram showing a processing flow of a general mapping update process.

FIG. 11 is a diagram showing a management main screen 500.

FIG. 12 is a diagram showing a file system browse screen 510.

FIG. 13 is a diagram showing a virtual storage area browse screen 520.

FIG. 14 is a diagram showing a physical storage device browse screen 540.

FIG. 15 is a diagram showing a data mapping change display screen 580.

FIG. 16 is a flowchart showing a basic processing procedure of data mapping change processing.

FIG. 17 is a flowchart showing a processing procedure for creating a data mapping change processing procedure plan.

[Code description] 40 ... I / O path, 46 ... network, 50 ... storage device, 54 ... disk device, 60 ... virtualization switch, 70 ... host, 72. .. Data mapping management server.