JP4115093B2 - Computer system - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a technique for rearranging data stored in a storage device, and particularly to a technique suitable for data rearrangement in a computer system having a plurality of storage devices.
[0002]
[Prior art]
As a technique for rearranging data stored in a storage subsystem in a computer system, there is a disk array system described in JP-A-9-274544. Here, the storage subsystem refers to a storage device system composed of a plurality of storage devices.
[0003]
A disk array system is a system that reads / writes data stored separately in each disk device by arranging multiple disk devices in an array and operating each disk device in parallel. . D. A. Patterson, G. Gibson, and R. H. As described in Kats, “A Case for Redundant Arrays of Inexpensive Disks (RAID)” (in Proc. ACM SIGMOD, pp. 109-116, June 1988), Types of level 1 to level 5 are given according to the redundant configuration. In addition to these types, a disk array system without redundancy may be called level 0.
[0004]
The cost for configuring the disk array system, the performance and characteristics of the disk array system, and the like vary depending on the level of the disk array system. Therefore, depending on the purpose of use of the system, a plurality of arrays (sets of disk devices) having different levels are often mixed when the disk array system is constructed. In such a disk array system in which arrays of different levels are mixed, the array of each level is called a parity group. Also, since the cost of disk devices also varies depending on performance, capacity, etc., a plurality of types of disk devices with different performance and capacity may be used in order to construct a system that realizes optimal cost performance.
[0005]
In a disk array system, since data is distributed and arranged in a plurality of disk devices, a host computer connected to the disk array system associates a logical storage area with a physical storage area indicating a storage area of the disk device ( Address conversion).
[0006]
In JP-A-9-274544, in a disk array system, data rearrangement between physical storage areas is executed, and the correspondence between the logical storage area of data and the physical storage area is indicated by the physical storage before the rearrangement. A technique for changing from an area to a physical storage area after rearrangement is disclosed. There is also a technology in which the disk array system manages the load status due to access to each logical storage area from the host computer, and determines the contents of the relocation so that the data is properly arranged after the relocation according to the actual results. It is disclosed.
[0007]
As a technology for transferring data between storage subsystems such as a host computer and a disk array system, M.M. T. There is a technique disclosed in O′Keefe, “Shared File Systems and Fiber Channel” (in Proc. Sixth Goddard Conference on Mass Storage Systems and Technologies, pp. 1-16, March 1998).
[0008]
This technology is a storage environment that connects multiple host computers and multiple storage subsystems via Fiber Channel (hereinafter referred to as “FC”), which is a high-speed network, and realizes data sharing via FC, a so-called Storage Area. This is a technology for realizing Network (SAN). In this way, by transferring data via FC, the load on the host computer and the network is reduced as compared with a general network.
[0009]
Network File System (NFS) is widely known as a technology for sharing data such as files held in a storage subsystem connected to a general network that does not use high-speed FC among multiple computers. Yes.
[0010]
When sharing data between networks using NFS, the load on the computer sharing the file and the network connecting the computer and the storage subsystem is greater than when using FC. However, if NFS is used, the existing network can be used. Therefore, compared to installing a new FC network, there are advantages such as reduced new equipment costs and easy management of file sharing. .
[0011]
[Problems to be solved by the invention]
As described above, with the technique disclosed in Japanese Patent Laid-Open No. 9-274544, data can be rearranged in one storage subsystem. However, in a computer system having a plurality of storage subsystems, data cannot be relocated between different storage subsystems. In addition, since the disk array system cannot recognize a file, data cannot be rearranged in units of files.
[0012]
On the other hand, using SAN technology enables high-speed data transfer in different storage subsystems. However, in the prior art, data could not be relocated between storage subsystems using SAN technology.
[0013]
In other words, in the prior art, the user who uses the host computer or system obtains information necessary for determining the proper arrangement of data, such as the load status of each storage area of each storage subsystem that constitutes the SAN. Nothing is thought about. For this reason, it cannot be determined how the host computer or the user can rearrange the data to realize the proper data placement.
[0014]
Furthermore, even if the user tries to relocate the data between the storage subsystems himself, the user has to perform detailed management of all unused areas to relocate the data. There was a heavy burden on the user.
[0015]
In addition, when data is transferred between different storage subsystems, the location of the data recognized by the application, that is, the access destination to be specified for the application to access the same data, changes before and after the relocation. The change is not considered at all in the prior art.
[0016]
Furthermore, there are the following problems when data is shared between computers connected by a general network using NFS.
[0017]
In other words, when a host computer (NFS server) used to realize data sharing by NFS has conventionally managed multiple storage subsystems, the NFS server itself is used between the multiple storage subsystems. Physical relocation of data could not be performed. For this reason, it was not possible to perform further distinction and management of storage areas of the storage subsystem, such as changing the physical location of shared data for each computer that shares data using an NFS server. .
[0018]
The present invention has been made in view of the above circumstances, and an object of the present invention is to enable a host computer including an NFS server to acquire information necessary for determining proper arrangement of data from a storage subsystem, and to create a SAN environment. Is to realize data rearrangement. Another purpose is to reduce the management burden on the user's storage subsystem. Another object of the present invention is to enable data relocation between different storage subsystems so that the position of data recognized by an application does not change before and after relocation. Furthermore, it is to enable rearrangement of data for a file.
[0019]
[Means for Solving the Problems]
  In order to solve the above-mentioned problems, the present inventionIn,A computer system having a computer and a plurality of storage subsystems connected to the computer, wherein each of the plurality of storage subsystems uses the storage area for each storage area of the storage subsystem. In accordance with an instruction from the computer, an acquisition unit that acquires a situation, a moving unit that moves data from one of the other storage subsystems, and notifies the computer that the migration is complete Notification means, and the computer collects the usage status of each storage area acquired by the acquisition means from each of the plurality of storage subsystems, and the storage for storing the data Maintains the correspondence between the area and the file managed by the file system executed on the computer The file correspondence recording unit, the storage status information of each of the plurality of storage subsystems collected by the collecting unit, and the file correspondence recording unit are used to store the information for each file. Based on the second collection means for collecting the usage status of the area and the usage status information for each file of the storage area of each of the plurality of storage subsystems collected by the second collection means, A second calculation means for calculating a storage area usage status in the system, a display unit for displaying at least the calculation result, and storage area usage status information of each storage subsystem collected by the collection means; Selection means for receiving an input of selection of each storage area displayed on the display unit from a user of the computer; A correspondence storage unit that stores a correspondence between a logical location of the data and a storage area in which the data is stored in the storage subsystem, and a notification that the data movement between the storage subsystems is completed by the notification unit And an update unit that updates the correspondence storage unit, and the instruction from the computer includes information specifying the storage area selected by the selection unit, It is characterized by that.
  Further, in the present invention, a computer system having a computer and a plurality of storage subsystems connected to the computer, wherein each of the plurality of storage subsystems is associated with each storage area of the storage subsystem. , Acquisition means for acquiring the usage status of the storage area, movement means for moving data from any of the plurality of other storage subsystems in accordance with instructions from the computer, and the movement has been completed Notification means for notifying the computer, the computer collecting from each of the plurality of storage subsystems the usage status of each storage area acquired by the acquisition means, A storage area for storing the data, and a file managed by a file system executed on the computer Using the file correspondence recording unit that holds the correspondence relationship, the storage area usage information of each of the plurality of storage subsystems collected by the collection unit, and the file correspondence recording unit A second collection unit that collects the usage status of the storage area for each file, and information on the usage status of each storage area of each of the plurality of storage subsystems collected by the second collection unit. Based on the second calculation means for calculating the usage status of the storage area in each storage subsystem, the calculated result, and the storage area information of each storage subsystem collected by the collection means A second selection unit that automatically selects the storage area that satisfies the following condition; a logical location of the data; and the storage subsystem. A storage unit that stores a correspondence with a storage area in which the data is stored in the storage system, and a notification that the data transfer between the storage subsystems is completed by the notification unit. Updating means for updating, and the instruction from the computer includes information specifying the storage area selected by the second selection means.
[0020]
Here, the usage status of each storage area is, for example, the usage status of the physical storage space of the storage area, the processing time consumed for access processing to the storage space, and the like.
[0021]
It is also conceivable that the unit of data moved by the storage subsystem and the unit of data managed by the computer are file units.
[0022]
Further, a configuration in which means for managing the entire logical storage area of all the storage subsystems and means for managing the correspondence between the characteristics of the storage device and the logical storage area is also conceivable.
[0023]
Further, a configuration in which a computer is added with means for sharing data stored in the storage subsystem between networks is also conceivable.
[0024]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a diagram showing a configuration of a first embodiment of a computer system to which the present invention is applied.
[0025]
The computer system of this embodiment includes a host computer (host) 100, a disk array 200, a switch 500, a client 800, and a local disk 190.
[0026]
The host 100 is connected to the client 800 and the disk array 200 via the network 700 by the network interface 170. The host 100 is also connected to the disk array 200 and the local disk 190 via the FC interface 160, the switch 500, and the FC 600.
[0027]
The host 100 has a file system 110, an operating system (hereinafter referred to as “OS”) 120, a manager 130, and an application 140 in a storage area of the host 100 itself.
[0028]
An application program (hereinafter simply referred to as an application) 140 issues a read or write request to the disk array 200 via the OS 120 and the file system 110.
[0029]
A general electronic computer is used as the host 100 and the client 800. Programs such as the file system 110 are stored in the local disk 190 outside the host 100, and are read into the host 100 and executed as necessary.
[0030]
When the host 100 has a storage device therein, it is also conceivable to use the storage device as the local disk 190.
[0031]
The local disk 190 stores various management information such as a logical unit (hereinafter referred to as “LU”) logical location name table 191 and an in-LU address logical location name table 195 used by the OS 120 and the file system 110. The LU logical location name table 191 holds information indicating the correspondence between the logical location name designated when the application 140 accesses the data of the disk array system 200 and the LU storing the data specified by the logical location name. It is a table.
[0032]
The in-LU address logical location name table 195 includes information indicating the correspondence between the logical location name specified when the application 140 accesses data in the disk array system 200 and the in-LU address of the data specified by the logical location name. It is a held table.
[0033]
The disk array 200 includes a control unit 300, a plurality of disk devices 210, a network interface 270, and an FC interface 260.
[0034]
The control unit 300 includes a CPU 310, a memory 320, and a cache memory 330 for executing processing.
[0035]
The memory 320 stores logical / physical correspondence information 321, a logical volume usage status 322, and a copy area management table 323. Details of such information will be described later.
[0036]
In this embodiment, an array (hereinafter referred to as “RAID”) is configured by n disk devices 210 (n is an integer of 2 or more), and the RAID by the n disk devices 210 is referred to as a parity group 220. Call.
[0037]
As RAID, redundant data (hereinafter referred to as “parity”) generated from the stored contents of n−1 disk devices 210 out of n disk devices 210 included in one parity group 220 are the remaining. A configuration of storing in one unit or a mirror disk (RAID level 1) configuration in which the content stored in n / 2 units is copied by the remaining n / 2 units is conceivable. Each parity group 220 can be regarded as one unit to be accessed from the host 100.
[0038]
In the present embodiment, attributes such as performance, reliability, and characteristics of each parity group 220 constituting the disk array 200 may be the same or different.
[0039]
The control unit 300 associates (address conversion) the logical volume read / written by the host 100 with the physical address indicating the storage area of the disk device 210, and provides the logical volume to the host 100. The disk array 200 can also be provided to the host 100 as a single LU by combining a plurality of logical volumes in address conversion. That is, the disk array 200 provides the host 100 with an LU composed of at least one logical volume. The host 100 reads / writes from / to the LU.
[0040]
In the present embodiment, physical rearrangement of data in consideration of the usage status of the disk array 200 between the plurality of disk arrays 200 is possible. Specifically, the disk array 200 acquires the usage status of the disk device 210 for reading / writing from the host 100. The host 100 collects the usage status acquired by each of the plurality of disk arrays 200 and presents it to the user. Further, the host 100 changes the LU logical position name table 191 in the local disk 190 in accordance with an instruction from the user who has received the usage status of the disk array 200, and the disk array 200 stores it in the LU. Copy the existing data. Thereby, LU rearrangement is performed among the plurality of disk arrays 200. The data can be rearranged in consideration of the usage status of the disk array 200, so that the data can be properly arranged.
[0041]
FIG. 2 shows a procedure of read / write processing performed by the disk array 200 in response to a read / write request from the host 100 and usage status acquisition processing in which the disk array 200 acquires the usage status of the disk device 210. FIG. The usage status acquisition process is performed at any time or when a read / write request is issued from the host 100.
[0042]
The application 140 of the host 100 designates a file by the file logical location name and requests the OS 120 to read / write the file. The OS 120 requests the file system 110 to read / write a file.
[0043]
The file system 110 accesses the local disk 190 via the FC interface 160, and obtains the LU number in which the designated file is stored from the LU logical position name table 191. The file system 110 obtains, from the LU address logical position name table 195, the LU address where the specified file is stored.
[0044]
The host 100 sends a small computer system interface (SCSI) standard read command or write command with an LU number or LU address to the disk array 200 that provides the LU of the calculated LU number via the FC interface 160. Issue.
[0045]
In a system in which the application 140 designates a file by describing the path to the logical location of the file using the logical drive name, directory name, and file name, the description of the path to the logical location (logical drive, directory, or file) This is the logical position name of the logical position. In general, the logical position name is information on a logical position that is used by an application to designate an access target.
[0046]
In order to manage each logical position, the file system 110 manages the hierarchical logical structure between the logical positions such as a directory structure, and the correspondence between the logical position name of each logical position and the LU number is the LU logical position name. It is described in the table 191 and managed. Also, the correspondence between the logical position name of each logical position and the address in the LU is described in the LU address logical position name table 195 and managed. The LU number also indicates the disk array 200 that provides the LU with that LU number (step 1000).
[0047]
The control unit 300 that has received the read / write command from the host 100 uses the logical / physical correspondence information 321 in the memory 320 to specify the logical volume that constitutes the LU specified by the LU number specified in the command. To do. The control unit 300 obtains an area of the disk device 210 corresponding to the logical volume, and converts the LU address specified by the command into a physical address.
[0048]
The logical / physical correspondence information 321 is a table that holds information about the correspondence between LUs and physical addresses of the disk devices 210, as shown in FIG.
[0049]
In the figure, an LU number 5001 and an LU address 5002 indicate the LU number and the LU address specified by the file system 110 of the host 100 in the read / write processing. The logical volume number 5003 is the number of the logical volume corresponding to the LU identified by the LU number 5001. The logical volume address 5004 is an address in the logical volume corresponding to the LU address 5002.
[0050]
The physical address is an address indicating an area on the disk device 210 in which data and parity are stored. The physical address has a parity group number 5005, a disk device number 5006 for each of data and parity, and an in-disk device address 5007. A parity group number 5005 indicates an individual parity group 220. A disk device number 5006 indicates an individual disk device 210. The disk device internal address 5007 is an address indicating an area in the disk device 210 (step 1010).
[0051]
In the case of data read, the control unit 300 reads data from the disk device 210 based on the physical address obtained by address conversion, and transfers it to the host 100. In the case of data writing, the control unit 300 stores the data transferred from the host 100 and the parity generated in association with the data at the physical address position of the disk device 210 obtained by the address conversion (step 1020).
[0052]
After completing the read / write process, the control unit 300 executes the usage status acquisition process. In this process, the control unit 300 determines the read / write type and sequential / random access type in the read / write process, and updates the logical volume usage status 322 of the logical volume that is the read / write target in the memory 320. . The logical volume usage status 322 is a table that holds information about the usage status of LUs included in the disk array 200. An example of the logical volume usage status 322 is shown in FIG.
[0053]
The logical volume usage status 322 describes the logical volume number 5101 and the disk usage time (in microsecond units) 5102 for each read / write type and sequential / random access type for each logical volume. Here, the time required for reading / writing is added to the disk usage time 5102 corresponding to the logical volume number 5101 of the logical volume to be read / written (step 1030).
[0054]
FIG. 5 is a flowchart showing the procedure of usage status collection processing in which the host 100 collects the usage status of the disk device 210 from each disk array 200. This process is performed at any time.
[0055]
The manager 130 of the host 100 issues a SCSI standard write command using the information collection parameter as write data to the command volume via the FC interface 160. The command volume is an information transfer LU included in the disk array 200 and is a logical volume in which a corresponding physical area is not designated. (Step 1100).
[0056]
Upon confirming that the issued command is a write command for the command volume, the control unit 300 determines the requested information from the operation code included in the information collection parameter transferred from the host 100. The controller 300 prepares the requested information on the memory 320 (step 1110). The control unit 300 reports the completion of writing to the host 100 via the FC interface 260 (step 1120).
[0057]
Upon receiving the completion report, the manager 130 of the host 100 issues a SCSI standard read command to the command volume of the disk array 200 via the FC interface 160 (step 1130).
[0058]
When receiving a read command for the command volume, the controller 300 transfers information prepared on the memory 320 to the host 100 via the FC interface 260 (step 1140). The controller 300 reports the completion of reading to the host 100 via the FC interface 260 (step 1150).
[0059]
The information collection parameter written in step 1100 and the information prepared in step 1110 include three types of information: logical volume information, parity group information, and usage status information.
[0060]
When the information collection parameter written in step 1100 is a logical volume information parameter as shown in FIG. 6, the control unit 300 is specified by the logical volume number 5201 specified by the 0th to 1st bytes. As shown in FIG. 7, logical volume information (information indicating the configuration of the logical volume in the disk array 200) is prepared.
[0061]
In the logical volume information shown in FIG. 7, various information 5202 of the logical volume specified by the logical volume number 5201 described in the 0th to 1st bytes is described in the 8th to 47th bytes. In the 49th to 121st bytes, information 5203 of each logical volume constituting the LU to which the logical volume belongs is described.
[0062]
When the information collection parameter is a parity group information parameter, the control unit 300 determines the parity group information of the parity group 220 to which the logical volume specified by the parameter belongs (such as the RAID configuration, the type name of the disk device 210, etc. Information indicating the configuration of the parity group 220 in the array 200 is prepared.
[0063]
When the information collection parameter is a parameter for confirming the usage status of the disk device 210, the control unit 300 uses the logical volume usage status information specified by the parameter (the usage status of resources in the disk array 200, for example, Information such as the time for which the logical volume is occupied, the number of received commands of the logical volume and the number of hits of the cache memory 330, the information of the occupied time of the processor 310 and the occupied time of the internal bus, etc.) are prepared.
[0064]
For each logical volume, the control unit 300 acquires the number of received commands, the number of hits of the cache memory 330, the occupation time of the processor 310, the occupation time of the internal bus, and the like in advance. For example, the manager 130 can obtain the occupation time rate per unit time by dividing the average of the occupation times acquired a plurality of times by the acquisition interval.
[0065]
The control unit 300 uses part or all of the logical / physical correspondence information 321 when generating logical volume information and parity group information. The manager 130 holds information regarding the performance of each type of disk device 210, and can obtain the performance of the disk devices 210 constituting the parity group 220 based on the type name of the disk device 210.
[0066]
Further, the manager 130 of the host 100 can obtain the logical volume number belonging to the LU from the response data by issuing a SCSI standard INQUIRY command to the LU and obtaining the response data.
[0067]
FIG. 8 is a flowchart showing the procedure of a rearrangement target determination process in which the host 100 determines data to be rearranged. This process is executed when an application used when a user searches for data to be rearranged is executed.
[0068]
The manager 130 of the host 100 determines the LU used by the OS 120 and the unused LU (free LU) from the LU logical location name table 191 stored in the local disk 190, for example. For each LU used by the OS 120, the manager 130 calculates the usage status of each logical volume in the disk array 200 to which the LU belongs, the usage status of the logical volume corresponding to the LU, and the like. For this calculation, the logical volume number belonging to the LU obtained by issuing the INQUIRY command, the logical volume information in each disk array 200 obtained by the usage status collection process, the parity group information, the usage status of the logical volume, and the like are used. (Step 1200).
[0069]
The manager 130 presents the calculation result such as the usage status to the user together with the attributes of the parity group 220 to which each logical volume belongs (RAID configuration, type name of the disk device 210, performance of the disk device 210, etc.) (step 1210). .
[0070]
For each LU, the manager 130 uses the logical volume number belonging to each LU obtained by issuing the INQUIRY command, logical volume information, parity group information, and logical volume usage in each disk array 200 obtained by the usage status collection process. Based on the status and the like, the usage status and the like of each logical volume corresponding to each free LU is calculated (step 1220). The calculation result is classified and presented to the user together with the attribute of the parity group 220 related to each free LU (step 1230).
[0071]
Information such as usage status can also be displayed on the host 100 or another computer connected to the host 100 via a network.
[0072]
The user refers to information about each LU of each disk array 200 and determines an LU (relocation source LU) to which data is to be relocated and a data relocation destination LU. However, instead of the user, the manager 130 of the host 100 may automatically determine the data rearrangement source or the rearrangement destination from the information about each LU. Determination of relocation is, for example, load distribution between disk arrays 200, load distribution between parity groups 220, allocation of LUs having files requiring high performance to high performance parity groups 220 after relocation, etc. Is performed so that is realized. The size of the relocation destination LU must be greater than or equal to the size of the relocation source LU. The size of each LU can be acquired by a SCSI standard READ CAPACITY command (step 1240).
[0073]
FIG. 9 is a flowchart showing a procedure of data rearrangement processing performed by the host 100. The host 100 executes this processing when there is an instruction from the user who has decided the rearrangement, for example, an execution command for instructing the rearrangement.
[0074]
The manager 130 of the host 100 to which the instruction from the user is input instructs the file system 110 to lock the relocation source LU (step 1300). In response to the lock instruction, the file system 110 prohibits acceptance of a read / write request to the relocation source LU (step 1310).
[0075]
Next, the manager 130 instructs the file system 110 to flush the cache memory for the relocation source LU (step 1320). The file system 110 stores data stored in the relocation source LU, which is cached in the memory on the host 100 and not yet written to the disk array 200, in the relocation source LU of the disk array 200. Writing is performed (step 1330).
[0076]
The manager 130 instructs the file system 110 to invalidate the cache for the relocation source LU (step 1340). The file system 110 invalidates the data stored in the relocation source LU and cached in the memory on the host 100 (step 1350).
[0077]
LU lock, cache memory flush, and invalidation processing correspond to LU unmount processing.
[0078]
The manager 130 instructs the disk array 200 where the relocation destination LU exists to copy data from the relocation source LU to the relocation destination LU. As in the usage status collection process, this instruction is a write including a copy instruction operation code, a copy instruction parameter such as a rearrangement source LU and a rearrangement destination LU in the command volume of the disk array 200 where the rearrangement destination LU exists. This is done by issuing a command (step 1360). The disk array 200 starts a copy process, which will be described later, and notifies the manager 130 that a copy instruction has been received (step 1370).
[0079]
The manager 130 rewrites the LU logical location name table 191 used by the file system 110 stored in the local disk 190, and swaps the logical location names of the relocation source LU and the relocation destination LU. An example of the LU logical position name table 191 to be replaced is shown in FIGS.
[0080]
In the figure, the disk array number, ID, and LUN are information necessary for specifying the LU number 6001. FIG. 10 shows the logical location name in the directory format, and FIG. 11 shows the logical location name in the drive format. Both indicate the logical position of the LU as a storage area used by the application 140 (step 1380). The manager 130 instructs the file system 110 to update (reread) the LU logical position name table 191 and release the lock instructed in Step 1300 (Step 1390).
[0081]
The file system 110 rereads the LU logical location name table 191 and updates the information (step 1400). The file system 110 releases the lock and resumes accepting the read / write request (step 1410).
[0082]
  Step 1400 and1410This process corresponds to the LU mounting process.
[0083]
  Step1410After the above process is executed, if the LU to be read / written in the file system 110 is an LU to be rearranged, the information in the read / write process of the file system 110 is replaced in step 1380. This is performed for the relocation destination LU.
[0084]
FIG. 12 is a flowchart showing a copy process procedure performed when the disk array 200 receives a copy instruction from the host 100 in the rearrangement process.
[0085]
When the disk array 200 in which the relocation destination LU exists receives a copy instruction from the host 100 via the FC interface 260, the control unit 300 stores the copy area management table 323 for the relocation destination LU specified by the copy instruction. Prepare on the memory 320.
[0086]
FIG. 13 shows the contents of the copy area management table 323. The copy area management table 323 is a table in which information such as the range and size of data to be copied is registered.
[0087]
In the figure, a copy destination LU number 6101 and a copy source LU number 6102 are areas for storing numbers that uniquely indicate the relocation destination LU and the relocation source LU in the network of the FC 600. More specifically, an 8-byte number (WORLD WIDE NAME), a 3-byte number (N_PORT ID), and a SCSI standard target ID or LUN specified by the host 100 as copy instruction parameters are stored. The copy block number 6103 is the number of blocks (minimum read / write unit) of the area to be copied, and stores data indicating the size of the copy area. Each block of the LU copy target area is assigned to the bit of the bitmap 6104. When the bit is “1”, it indicates that the copy has not been made, and when it is “0”, it indicates that the copy has been completed. At the initial stage, all bits corresponding to the copy target area are set to 1 (step 1500).
[0088]
The control unit 300 notifies the host 100 of receipt of the copy instruction. This notification is performed at the time after the copy instruction is actually received and after the copy area management table 323 is set and before the actual copy is performed. For this reason, the time from receipt of the copy instruction to the notification is short (step 1510).
[0089]
The control unit 300 reads data to be stored from the relocation source LU via the FC interface 260, and performs copying to be stored in the relocation destination LU (step 1520).
[0090]
The control unit 300 sequentially changes the bit of the bitmap 6104 corresponding to the copied area to 0 for the LU copy target area (step 1530). When the copying of the entire target LU is completed, the control unit 300 ends the copy process (step 1540).
[0091]
When the disk array 200 in which the relocation source LU exists and the disk array 200 in which the relocation destination LU exists are the same, the LU is copied in the disk array 200.
[0092]
Read / write from the host 100 to the relocation target LU is performed on the relocation destination LU, that is, the disk array 200 in which the relocation destination LU exists, even if the data of the relocation target LU is being copied. .
[0093]
FIG. 14 shows a processing procedure when the disk array 200 in which the relocation destination LU exists receives a read / write command for the LU to be relocated during the copy processing in the data relocation. FIG.
[0094]
When the disk array 200 receives a read command via the FC interface 260, the control unit 300 compares the read target range with the bitmap 6104 of the table 323 (step 1610). If the read target area includes an uncopied area, the control unit 300 preferentially reads the data in the read target area and completes the copy (step 1630). The control unit 300 updates the bit corresponding to the read target area of the bitmap 6104 to be copied (step 1640). The control unit 300 transfers the copied data in the disk array 200 to the host 100 (step 1650). If all the read target areas have been copied, the control unit 300 transfers the copied data in the disk array 200 to the host 100 (step 1650).
[0095]
When receiving the write command via the FC interface 260, the control unit 300 writes the data transferred from the host 100 to the write target area (step 1670). The control unit 300 updates the bit corresponding to the write target area of the bitmap 6104 in the copy area management table 323 to be copied (step 1680). The control unit 300 continues to copy the remaining uncopied area (step 1690).
[0096]
With the above processing, the disk array 200 in which the rearrangement destination LU exists can process read / write from the host 100 even during copy processing in data rearrangement.
[0097]
At the time of this read / write process, the control unit 300 also performs the usage status acquisition process described above.
[0098]
In addition, the manager 130 of the host 100 issues a data write command including parameters for acquiring copy progress to the command volume of the disk array 200 during the copy processing in the data rearrangement, and the data prepared by the disk array 200 is stored. By reading, the disk array 200 can be inquired about copy progress information and the like.
[0099]
Specifically, the control unit 300 that has received a write command for the command volume confirms the parameters written to the command volume. The control unit 300 refers to the copy area management table 323, prepares information such as the copy progress rate corresponding to the parameter on the memory 320, and notifies the host 100 of the write completion. The manager 130 reads the command volume. The control unit 300 responds to an inquiry about the progress of copying by transferring data prepared on the memory 320 in response to the read of the host 100.
[0100]
According to the present embodiment, the proper arrangement of data by relocation of LUs among a plurality of disk arrays 200 is logically equivalent to the application 140 before and after the relocation, that is, the application can access the access to be accessed. This can be realized while the logical position name to be used does not change.
[0101]
Further, according to this aspect, the computer determines the appropriate arrangement of data from the viewpoint of, for example, load distribution of storage device resources, etc., based on the physical storage device resource usage status of each storage area acquired from the disk array. Can be used. Therefore, using this information, for example, data can be properly arranged by rearranging data between different storage subsystems.
[0102]
In the present embodiment, the data rearrangement between the plurality of disk arrays 200 has been described. However, the storage subsystem that stores the relocation target data may not be a disk array subsystem. Another type of storage subsystem using a magnetic disk device, a magneto-optical disk device, a magnetic tape device, or a semiconductor disk device may be used.
[0103]
Note that the manager 130 may collect information and give instructions via the network 700, for example, using a protocol defined by Simple Network Management Protocol (SNMP) instead of via the FC 600.
[0104]
In this embodiment, the case where the logical volume usage status 322 acquired by the control unit 300 of the disk array 200 is a cumulative value of usage time has been described. However, the control unit 300 may accumulate the usage time per unit time in the form of usage rate in the memory 320 and collect it as the logical volume usage status 322 and collect it by the manager 130 of the host 100.
[0105]
FIG. 15 is a diagram showing a configuration of a second embodiment of a computer system to which the present invention is applied.
[0106]
As shown in the figure, the computer system of this embodiment has a configuration in which an LU area range table 192 is stored in the local disk 190 and a copy control unit 510 is provided in the switch 500. Different from the computer system.
[0107]
In this embodiment, the disk array 200 acquires the usage status of the disk device 210, the host 100 collects the usage status from the plurality of disk arrays 200, and presents the usage status to the user including analysis based on the file of the computer system. To do. The host 100 changes data for file management (hereinafter referred to as “metadata”). The switch 500 copies data stored in the disk array 200 based on an instruction from the host 100. As a result, the files can be rearranged among the plurality of disk arrays 200, and the data can be properly arranged.
[0108]
In the first embodiment, the file system 110 of the host 100 manages each LU by distinguishing between LUs in use and those not in use. In this embodiment, the file system 110 uses all LUs, and manages a set of all LU areas as a single area (hereinafter referred to as “integrated area”). Further, the files on the integrated area are managed by metadata described later. The metadata is stored at a predetermined position in the integrated area.
[0109]
FIG. 16 is a diagram showing an example of the LU area range table 192 used by the file system 110 to manage the integrated area in this embodiment. The LU area range table 192 holds information indicating the correspondence between the range of the integrated area and the range of each LU area.
[0110]
In the drawing, an in-area address 6301 stores an address in the integrated area. The LU number 6302 includes a disk array number, ID, and LUN, and indicates an LU stored in the area address 6301. The LU address 6303 stores an address in the LU specified by the corresponding LU number 6302.
FIG. 17 is a flowchart showing a processing procedure when the host 100 performs read / write.
[0111]
It is assumed that the application 140 of the host 100 reads and writes data stored in the disk array 200 by specifying a logical position of a file managed by the file system 110. The file system 110 stores metadata in the disk array 200 in order to manage data as files.
[0112]
The metadata is stored in the disk array 200, but may be cached in the memory on the host 100 based on the management of the file system 110. Hereinafter, a case where metadata is cached in a memory on the host 100 will be described.
[0113]
FIG. 18 is a diagram showing the contents of metadata.
[0114]
As shown in the figure, the metadata includes the creation date / time, update date / time, access date / time, attribute, file logical position name, security information, file position, and the like of each file. The range in the integrated area corresponding to each file is indicated by information stored in the file position 6407.
[0115]
The application 140 of the host 100 designates the file by the file logical position name and requests the OS 120 to read / write the file (Step 1700). The OS 120 requests the file system 110 to read / write the file (step 1710). The file system 110 refers to the metadata stored in the cache memory, and obtains the position of the designated file (LU and LU address) from the metadata and information in the LU area range table 192 (step 1720).
[0116]
If the request is a write request, the file system 110 further updates the metadata (step 1740). The file system 110 performs read / write in the area indicated by the file position obtained in step 1720 to the disk array 200 (step 1750), and updates the cached metadata and the metadata of the disk array 200 (step 1750). Step 1760).
[0117]
The metadata update in steps 1740 and 1760 stores the accessed file in the creation date and time 6401, update date and time 6402, access date and time 6403, attribute 6404, file logical location name 6405, security information 64060, file location 6407, and the like. This information is updated by updating the information according to the access contents. For example, when the file size increases or decreases due to writing, the range in the area indicated by the metadata file position 6407 is increased or decreased accordingly. When a new file is created, an entry for the new file is added to the metadata. When a file is deleted, the corresponding entry is deleted.
[0118]
In the present embodiment, the control unit 300 performs the same usage status acquisition process as in the first embodiment. Further, the manager 130 of the host 100 performs the same usage status collection process as in the first embodiment.
[0119]
FIG. 19 is a flowchart showing a procedure of file-based relocation target determination processing performed by the host 100.
[0120]
The manager 130 of the host 100 inquires the file system 110 about the correspondence between the file and the LU for each file existing in the integrated area (step 1800). The file system 110 answers the query using the cached metadata and the LU area range table 192 (step 1810).
[0121]
The manager 130 calculates the usage status of each logical volume for each disk array 200, the usage status of each logical volume of each LU, the usage status of each logical volume for each file, and the like. For this calculation, the logical volume number belonging to each LU in each disk array 200 obtained by the INQUIRY command, the logical volume information and parity group information in each disk array 200 obtained by the usage status collection process, and the usage status of the logical volume Are used (step 1820). The manager 130 presents the calculation result to the user together with the attribute of the parity group 220 to which each logical volume belongs. That is, the host 100 provides the user with information on the usage status from various viewpoints such as the disk array 200, logical volume, LU, and file (step 1830).
[0122]
The manager 130 calculates available space for the LU and logical volume provided by each disk array 200 and presents it to the user (step 1840). The manager 130 inquires of the file system 110 about available space for the LU and logical volume provided by each disk array 200 (step 1850). The file system 110 refers to the cached metadata and the LU area range table 192, identifies a free area where no file exists, and responds to the manager 130 (step 1860). The manager 130 classifies and presents the usage status of the logical volume in the free area together with the logical volume and the attribute of the parity group 220 from the various usage statuses obtained by the usage status collection processing (step 1870).
[0123]
The usage status and free space information can be displayed on the host 100 or another computer connected to the host 100 via a network. The user determines a file to be rearranged and a relocation destination free area from these pieces of information. The manager 130 may automatically determine the same relocation target and free space from these pieces of information (step 1880).
[0124]
The file system 110 of the host 100 may monitor the read / write request frequency (access frequency) to each file from the OS 120 or the application 140 to generate statistical information, and present it to the user in step 1830. .
[0125]
Thereby, the user can determine the file to be rearranged in consideration of the access frequency of each file in the host 100.
[0126]
FIG. 20 is a flowchart showing the procedure of the rearrangement process performed by the host 100 in response to the result of the rearrangement target determination process. This process is basically the same as the process of replacing the LU as a file and the disk array 200 as the switch 500 in the procedure of the LU unit rearrangement determination process shown in FIG. Only the parts different from FIG. 9 will be described below.
[0127]
The manager 130 instructs the file system 110 to reserve the use of an area for the relocation destination free area (step 1940). The file system 110 updates the cached metadata so that the designated relocation destination area is secured (step 1950). The manager 130 instructs the file system 110 to flush the cache memory of metadata (step 1960). The file system 110 writes the metadata cached in the memory on the host 100 to the disk array 200 (step 1970).
[0128]
The manager 130 rewrites the metadata and changes the position of the designated file from the rearrangement source area to the rearrangement destination area. As a result, the relocation source area is set as an empty area (step 2010). The manager 130 instructs the file system 110 to invalidate the cache for the metadata (step 2020). The file system 110 invalidates the metadata cached in the memory on the host 100 (step 2030).
[0129]
Thereafter, when the file system 110 reads / writes to the file, it can normally read / write the data copied to the relocation destination area.
[0130]
According to the present embodiment, it is possible to appropriately arrange the files among the plurality of disk arrays 200 so that the applications 140 are logically equivalent before and after the relocation.
[0131]
FIG. 21 is a diagram showing a configuration of a third embodiment of a computer system to which the present invention is applied.
[0132]
In the computer system of this embodiment, the client 800 includes an FC interface 860 and a network interface 870. The client 800 is connected to the host 100, the disk array 200, and the switch 500 via the FC interface 860 via the FC 600, and is connected to the host 100 and the disk array 200 via the network 700 via the network interface 870. This is different from the computer system of the second embodiment. In the present embodiment, a plurality of clients 800 and the host 100 share files on the disk array 200. The client 800 has an OS 820 and an application 840. The client 800 is a general electronic computer.
[0133]
Similar to the second embodiment, the file system 110 according to the present embodiment uses all LUs, collects all LU areas, and manages them as a single integrated area. Then, the files on the integrated area are managed by metadata as in the second embodiment.
[0134]
A process in which the client 800 accesses a file stored in the disk array 200 will be described.
[0135]
FIG. 22 is a flowchart showing a processing procedure when the client 800 reads a file stored in the disk array 200.
[0136]
The application 840 of the client 800 requests the OS 820 to read the file (Step 2100). The OS 820 notifies the file system 110 of the host 100 of the file read via the network interface 870 or the FC interface 860 (step 2110).
[0137]
Upon receiving the file read notification, the file system 110 obtains the LU in which the file is stored and the address within the LU with reference to the metadata and the LU area range table 192 (step 2120). The file system 110 locks the LU address of the LU in which the file is stored against the write from the other client 800 (step 2130). The file system 110 flushes the metadata in the cache memory of the host 100 (Step 2140). The file system 110 returns to the OS 820 of the client 800 the LU and LU address where the file is stored, and the LU and LU address where the metadata is stored (step 2150).
[0138]
Upon receiving the response, the OS 820 of the client 800 reads, via the FC interface 860, the read within the LU in which the file is stored, to the disk array 200 in which the LU in which the file to be read is stored exists. Go to process request from application 840 (step 2160).
[0139]
When the read processing of the data requested from the client 800 is completed, the OS 820 updates the access date and time of the file on the metadata at the LU and the LU address notified from the file system 110 of the host 100 (step 2170). The OS 820 notifies the file system 110 of the completion of processing via the network interface 870 or the FC interface 860 (step 2180).
[0140]
Upon receiving the completion notification, the file system 110 invalidates the metadata cache memory on the host 100 (step 2190), and releases the lock performed in step 2130 (step 2200).
[0141]
FIG. 23 is a flowchart showing a processing procedure when writing is performed.
[0142]
The write process is almost the same as the process of replacing the read with the write in the read process of FIG. Hereinafter, different parts will be described.
[0143]
The file system 110 describes an area reservation for a file use area that may increase due to writing in the metadata (step 2340). In the file system 110, the LU in which the file is stored, the address in the LU (including the area reserved for the file usage area that may increase by writing), and metadata are stored in the OS 820 of the client 800. Returns the current LU and the LU address. It is assumed that the increase amount of the file usage area that may increase due to the write is included in the write notification from the OS 820 of the client 800 (step 2360).
[0144]
Upon receiving the response, the OS 820 writes to the address in the LU in which the file is stored via the FC interface 860 to the disk array 200 in which the LU in which the file to be written is stored exists, and the application The request from 840 is processed (step 2370).
[0145]
In this way, by processing the access of the client 800, the client 800 and the host 100 can share and use the files stored in the disk array 200 without contradiction. The file access of the host 100 itself is processed in the same manner as the file access by the client 800.
[0146]
Next, file rearrangement in the present embodiment will be described.
[0147]
The processing related to file relocation in this embodiment (usage status acquisition processing, usage status collection processing, relocation target determination processing, and relocation processing) is the same as that in the second embodiment. However, the relocation process is not executed while the file is locked in the data read / write process requested by the application 840. In addition, the flushing of the file cache memory and the writing back to the disk array 200 in steps 1920 and 1930 of the relocation processing shown in FIG. 20 are performed for the client 800 in which the file system 110 caches the file. Instruct them to do this.
[0148]
According to the present embodiment, even in an environment where data stored in the disk array 200 is shared and used, physical rearrangement of files between the plurality of disk arrays 200 can be performed with respect to the applications 140 and 840. It can be performed so as to be logically equivalent before and after the rearrangement.
[0149]
Also in this embodiment, the file system 110 of the host 100 monitors the read / write request frequency to each file from the OSs 120 and 820 and the applications 140 and 840, generates statistical information, and performs relocation target determination processing. You may make it show to a user.
[0150]
In this embodiment, a program of the manager 130 is stored on the client 800, and the manager 130 uses the FC interface 860 or the network interface 870 to collect information such as usage status and perform processing such as instructions. The request may be made to the file system 110 or the disk array 200.
[0151]
FIG. 24 is a diagram showing the configuration of the fourth embodiment of the computer system to which this invention is applied.
[0152]
The computer system of this embodiment differs from the computer system of the first embodiment in that the host 100 has an LU pool manager 900 and an LU management table 910.
[0153]
According to the present embodiment, it is possible to easily select the LU relocation destination.
[0154]
FIG. 25 is a diagram showing the LU management table 910.
[0155]
The LU management table 910 is a table in which information related to the LU status of the entire system is registered.
[0156]
In the LU number 3310, a number uniquely assigned to each LU is registered. This number is used by the LU pool manager 900 to manage each LU. In the size 3320, the capacity of the corresponding LU is registered. The configuration 3330 stores the type of RAID configuration. In the configuration 3330, when the LU is configured by a cache memory 330 or a single disk, information about the LU is also stored.
[0157]
The status 3340 stores information indicating the LU status. As types thereof, “online”, “offline”, “unimplemented”, and “failure offline” are provided. “Online” indicates that the LU is in a normal state and is accessible from the host 100. “Offline” indicates that an empty LU, that is, an LU exists normally, but is inaccessible from the host 100. “Not mounted” indicates that this LU is not defined and cannot be accessed from the host 100. “Failure offline” indicates that a failure has occurred in the LU and access from the host 100 is not possible.
[0158]
The disk array number 3350 stores information indicating the disk array 200 in which the corresponding LU exists.
[0159]
In the path 3360, a number indicating which of a plurality of FCs 600 connected to each disk array 200 is assigned an LU is stored. In the ID 3370 and the LUN 3380, numbers indicating LUs are stored.
[0160]
  The disk performance 3390 stores an index indicating the performance of the disk device 210 in which the corresponding LU is currently arranged. In particular,FIG.As shown in FIG. 4, the performance of the disk device 210 is classified into high performance, medium performance, and low performance indexes from the average seek time, average rotation waiting time, and configuration of the disk device 210 and stored. The LU on the cache memory is classified as ultra-high performance.
[0161]
The emulation type 3400 stores information indicating the type of each LU provided by the disk array 200 to the host 100 as a disk device.
[0162]
The relocation enable flag 3410 stores a flag for designating whether or not the LU can be used as a relocation destination when performing LU relocation. The user can distinguish the relocation LU from other LUs using this flag. The user can change on / off of the flag.
[0163]
FIG. 25 shows an LU management table for disk array number 0. The manager 130 holds LU management tables for all the disk arrays 200.
[0164]
The determination of the rearrangement target in the present embodiment is performed as follows.
[0165]
The user designates the relocation source LU and the conditions required as the relocation destination LU to the manager 130. Specific conditions include performance conditions and reliability levels.
[0166]
For example, if a certain LU is used excessively and the load exceeds the capacity of the disk device that includes the LU, the processing capacity of the LU can be specified by specifying a higher-performance disk device as the relocation destination of the LU. Can be expected to improve the performance of the computer system.
[0167]
Further, when an LU storing important data exists on a single disk or a non-redundant RAID (RAID 0), fault tolerance due to redundancy can be ensured by specifying RAID 5 or RAID 1 as the relocation destination.
[0168]
The manager 130 uses the information registered in the LU management table 910 to determine the relocation destination LU, notifies the user, and relocates the LU.
[0169]
FIG. 26 is a flowchart showing the procedure of the rearrangement target determination process in the present embodiment. This process is executed in response to a user instruction.
The user designates the disk array number, path, ID, and LUN of the relocation source LU to the manager 130. In this case, a disk array number and LU number may be specified instead of the path and ID (step 2500).
[0170]
The user designates performance conditions and reliability levels as requirements for the relocation destination to the manager 130 (step 2510).
[0171]
The manager 130 notifies the LU pool manager 900 of the request conditions for the relocation source LU and the relocation destination (step 2520). The LU pool manager 900 searches the LU management table 910 and checks whether there is an LU that satisfies the requested condition (step 2530).
[0172]
In this case, the search conditions are: “state is offline”, “size is greater than or equal to relocation source LU”, “emulation type is the same as relocation source LU”, and “relocatable flag is on (true) or possible” and “ The performance condition must meet the requirement "and the reliability level must meet the requirement.
[0173]
If there is an LU that satisfies the condition in step 2540, the LU pool manager notifies the manager 130 of the corresponding LU (step 2550). The manager 130 determines the notified LU as a relocation destination LU (step 2560).
[0174]
If there is no LU satisfying the condition in step 2540, the LU pool manager 900 searches the LU management table 910 for the LU number 3310 of “status not mounted” (step 2570).
[0175]
If there is no unmounted LU number 3310, the LU pool manager 900 notifies the manager 130 that an LU that satisfies the condition cannot be used (step 2580). Upon receiving the notification, the manager 130 notifies the user that the relocation destination LU cannot be determined (step 2590).
[0176]
If there is an unmounted LU in step 2570, the LU pool manager 900 specifies the conditions for the unmounted LU number and the relocation destination LU, and constructs the relocation destination LU in the corresponding disk array 200. Instruct (step 2600).
[0177]
In this case, the conditions for the relocation destination LU are “size is equal to or larger than the relocation source LU”, “emulation type is the same as the relocation source LU”, “performance condition meets requirements”, and “reliability level is required” Satisfy.
[0178]
The disk array 200 instructed to construct the LU performs LU construction processing (step 2610). If the construction is successful, the disk array 200 notifies the LU pool manager 900 of a series of information including the disk array number, path, ID, LUN, etc. for the constructed LU (step 2620). If the construction fails, the disk array 200 notifies the LU pool manager 900 that the construction is impossible (step 2610).
[0179]
The LU pool manager 900 registers the notified LU information in the LU management table 910 (step 2630) and notifies the manager 130 (step 2550). The manager 130 determines this LU as a relocation destination LU (step 2560).
[0180]
The LU pool manager 900 that has received the notification that the construction is not possible notifies the manager 130 that the LU that satisfies the condition cannot be used (step 2580). Upon receiving the notification, the manager 130 notifies the user that the relocation destination LU cannot be determined (step 2590).
[0181]
FIG. 27 is a flowchart showing the procedure of LU construction processing performed by the disk array 200. This processing is performed when an instruction from the LU pool manager 900 is received.
The disk array 200 receives the condition regarding the unmounted LU number and the relocation destination LU in accordance with an instruction from the LU pool manager 900 (step 2700).
[0182]
The disk array 200 compares the received conditions with internal resource allocation statuses such as the disk device 210 and the cache memory 330, and determines whether or not an LU with the requested conditions can be constructed (step 2710). If an LU can be constructed, the disk array 200 allocates internal resources and performs formatting / initialization processing to construct the LU. The disk array 200 assigns an LU number corresponding to an unmounted LU received from the LU pool manager 900 to the constructed LU (step 2720).
[0183]
The disk array 200 sets the FC interface 260 and assigns a path, ID, and LUN to the LU (step 2730). The disk array 200 notifies the LU pool manager 900 of a series of information about the constructed LU, including the disk array number, path, ID, LUN, and the like (step 2740).
[0184]
If the LU cannot be constructed in step 2710, the disk array 200 notifies the LU pool manager 900 that construction is impossible (step 2750).
[0185]
When the relocation destination LU is determined, the manager 130 performs relocation processing for the relocation source LU and the relocation destination LU, as in the first embodiment.
[0186]
FIG. 28 is a flowchart showing the procedure of offline processing of the relocation source LU.
[0187]
The manager 130 acquires the progress of copying by the method described in the first embodiment, and when copying is completed, instructs the LU pool manager 900 to take the relocation source LU offline (step 2800).
[0188]
The LU pool manager 900 that has received the offline instruction instructs the disk array 200 of the relocation source LU to take the relocation source LU offline (step 2810). The disk array 200 that has been instructed to go offline takes the LU offline by setting the FC interface 260 and locking the LU (step 2820). The disk array 200 notifies the LU pool manager 900 that it has gone offline (step 2830).
[0189]
The LU pool manager that has received the notification of offline update updates the contents of the LU status 3340 of the LU management table 910 to offline (2840).
[0190]
Here, an example in which the manager 130 acquires copy progress information has been described, but the disk array 200 may notify the manager 130 of the end of copying.
[0191]
Further, instead of the manager 130 instructing to go offline, the disk array 200 may take the relocation source LU offline when the copy is completed, and notify the LU pool manager 900 that it has gone offline.
[0192]
In the present embodiment, the case where the manager 130 issues a copy instruction to the copy control unit 510 of the switch 500 using the SCSI standard EXTENDED COPY command has been described, but other commands may be used. The other command is, for example, a write command to the command volume. Further, as shown in FIG. 15, the disk array 200 has a copy control unit 510, and the manager 130 instructs the copy control unit 510 of the disk array 200 to perform a copy instruction so that the disk array 200 performs a copy process. May be.
[0193]
In the present embodiment, it is assumed that the user specifies information such as conditions requested as the relocation destination LU, but the manager 130 may automatically determine and specify the information.
[0194]
In this embodiment, the LU pool manager 900 and the manager 130 exist on the same host 100, but the LU pool manager 900 may exist on a computer different from the manager 130, such as the remote computer 400. In this case, the LU pool manager 900 and the manager 130 give instructions and notifications via the FC 600 and the network 700 using SCSI, SNMP, other protocols, and command systems.
[0195]
According to the present embodiment, in LU relocation processing, it is possible to easily manage and select a relocation destination LU, reduce the load on the user, and facilitate management of the computer system.
[0196]
FIG. 29 is a diagram showing a fifth embodiment of a computer system to which the present invention is applied. In the computer system of this embodiment, the host 100 reads / writes a file based on a read / write request from the client 800 using the LU area range table 193 with new items added to the LU area range table 192. And the point which performs the process which transfers data between the clients 800 via the network 700 differs from the computer system of 3rd Embodiment.
Network File System (NFS) and Common Internet File System (CIFS) are widely used as protocols for file sharing via the network 700. By using these protocols and the widely used network 700, a file sharing environment can be easily realized. Also in this embodiment, it is considered to use NFS or CIFS.
[0197]
FIG. 30 is a diagram showing the LU area range table 193. The LU area range table 193 stores information regarding whether or not access to the LU area uses the network.
[0198]
In the usage type 3510, the LU area is an LU area in which read / write processing is performed via the FC 600 as in the third embodiment, or as described in this embodiment, read / write via the network 700 is performed. Information indicating whether the LU area is to be written is stored.
[0199]
In the usage type 3510, the LU area is an area used in the configuration and method for relocating the LU as in the first embodiment (in this case, the read / write request passes through the FC 600), or the read to the LU. Information indicating whether the / write processing is an area to be performed via the network 700 can be stored. In the usage type 3510, information on unused areas can also be stored. In addition, the in-area address, disk array number, ID, LUN, and in-LU address are the same as those described in the third embodiment, and a description thereof will be omitted.
[0200]
By centrally managing LUs using the LU area range table 193, the file system 110 can manage the LUs as a plurality of areas that are distinguished at least for each use type.
[0201]
By setting the LU area range table 193, the host 100 uses whether the request from the client 800 is an access by the method described in the third embodiment or an access through a network. Are distinguished by the protocol to be used. The host 100 treats the LU area separately for each usage type 3510 according to this type.
[0202]
The host 100 distinguishes between files and areas accessed by the method of the third embodiment and files and areas accessed by the method of the present embodiment. Therefore, methods for accessing the same file and area are not mixed.
[0203]
The host 100 also makes the same distinction when searching for accessible files. In other words, when there is an access request from the client 800 to each file existing in the same disk array 200, by identifying the usage type from the client 800, another usage type different from the usage type of the client 800 is obtained. The file is not returned to the client 800. Accordingly, the client 800 is notified only of the files that can be accessed only by the access method used by the client 800. This makes it possible to easily manage shared files in this system.
[0204]
Further, as in the first embodiment, there are an area used for the configuration and method for relocating the LU (read / write via the FC 600) and an area where read / write to the LU is performed via the host 100 and the network 700. By making the distinction, the effects as described above can be obtained for all of these types of use. Further, the user can freely set the LU area range table 193 via the host 100 or the remote computer 400.
[0205]
In the present embodiment, a file sharing protocol such as NFS or CIFS is used via the network 700, and data transfer between the host 100 and the client 800 is performed via the network 700, but instead of the network 700, via the FC 800. Processing to be performed is also conceivable. Furthermore, a process in which the client 800 performs a read / write request to each LU via the host 100 and the network 700 is also conceivable. In this case, the host 100 obtains the read / write target area requested by the client 800 using the LU area range table 192. The host 100 reads the target data and transfers it to the client 800 via the network 700. Alternatively, the host 100 receives data from the client 800 via the network 700 and writes the data.
[0206]
FIG. 31 is a flowchart showing the processing procedure of the host 100 when the application 840 of the client 800 reads a file stored in the disk array 200.
[0207]
As in the third embodiment, the file system 110 of the host 100 that has received the read notification refers to the LU area range table 193 and the metadata to obtain the LU and LU area in which the file is stored (step 2900). The file system 110 locks the file to be read in response to another write request (step 2910). The file system 110 reads the data in the file (step 2920) and transfers the read content to the client 800 via the network 700 (step 2930). The file system 110 updates the file access date and time on the metadata (step 2940). The file system 110 unlocks the file (step 2950), and notifies the client 800 of the completion of the read process (step 2960).
[0208]
FIG. 32 is a flowchart illustrating a processing procedure when the application 840 performs writing.
[0209]
Receiving the write notification, the host 100 receives write data from the client 800 via the network 700 (step 3000). The host 100 refers to the LU area range table 193 and the metadata to obtain the LU in which the file is stored and the LU internal area (step 3010). The host 100 locks the file (step 3020) and writes data to the file. At this time, if necessary, the metadata is updated and a file use area is added (step 3030).
[0210]
The host 100 updates the file update date and access date on the metadata (step 3040). The host 100 releases the lock (step 3050) and notifies the client 800 of the write completion (step 3060).
[0211]
FIG. 33 is a flowchart showing a processing procedure when the application 840 or the OS 820 of the client 800 inquires the host 100 about the existence of an accessible file.
[0212]
When there is a request from the application 840 or the OS 820 itself, the OS 820 inquires about the existence of a file accessible to the host 100 via the network 700 (step 3100).
[0213]
Upon receiving the notification, the file system 110 of the host 100 obtains an accessible file with reference to the LU area range table 193 and metadata (step 3110). The file system 110 notifies the client 800 of information such as the file name of each file (step 3120).
[0214]
In this embodiment, the client 800 and the host 100 can share and use the files stored in the disk array 200 via the host 100. The data rearrangement method and the like are the same as in the third embodiment. However, the rearrangement process is performed in each use type area.
[0215]
In the present embodiment, even in an environment where data stored in the disk array 200 is shared and used, the physical relocation of files between the plurality of disk arrays 200 is performed without the involvement of the application 140 and the application 840. It can be carried out.
[0216]
The present invention is not limited to each embodiment, and various modifications are possible within the scope of the gist.
[0217]
For example, as shown in FIGS. 1, 15, and 21, the manager 130 may be arranged outside the host 100 as a program on the remote computer 400 having a network interface 470 and an FC interface 460. A manager 130 external to the host 100 collects and instructs information via the FC 600 or the network 700, performs the same processing as in each embodiment, and performs proper data placement by relocation of LUs among a plurality of disk arrays 200. This can be done transparently to the application 140.
[0218]
In the first embodiment, file sharing or the like may be performed as in the third embodiment. Also in this case, as in the first embodiment, the physical rearrangement of data between the plurality of disk arrays 200 is transparent to the applications 140 and 840 so as to be logically equivalent before and after the rearrangement. Can be done.
[0219]
【The invention's effect】
As described above, according to the present invention, data can be easily rearranged between storage subsystems. Furthermore, according to the present invention, information necessary for the host computer to determine proper placement can be acquired from a plurality of storage subsystems. In addition, data relocation between different storage subsystems can be performed such that the data position for the application does not change before and after relocation. Furthermore, it is possible to perform data relocation for files among different storage subsystems.
[Brief description of the drawings]
FIG. 1 is a diagram showing a configuration of a computer system to which a first embodiment of the present invention is applied.
FIG. 2 is a flowchart showing a procedure of read / write processing and usage status acquisition processing in the first embodiment of the present invention;
FIG. 3 is a diagram showing logical / physical correspondence information used in the first embodiment of the present invention.
FIG. 4 is a diagram showing a usage status of a logical volume used in the first embodiment of the present invention.
FIG. 5 is a flowchart showing a procedure of usage status collection processing in the first embodiment of the present invention.
FIG. 6 is a diagram showing parameters of logical volume information used in the first embodiment of the present invention.
FIG. 7 is a diagram showing logical volume information used in the first embodiment of the present invention.
FIG. 8 is a flowchart showing a procedure of a rearrangement target determination process in the first embodiment of the present invention.
FIG. 9 is a flowchart showing a procedure of rearrangement processing in the first embodiment of the present invention.
FIG. 10 is a diagram showing an LU logical location name table used in the first embodiment of the present invention.
FIG. 11 is a diagram showing an LU logical location name table used in the first embodiment of the present invention.
FIG. 12 is a flowchart showing a copy processing procedure according to the first embodiment of the present invention;
FIG. 13 is a diagram showing a copy area management table used in the first embodiment of the present invention.
14 is a flowchart showing a processing procedure for a read / write command to a relocation destination LU during copying by the processing shown in FIG. 16 in the first embodiment of the present invention. FIG.
FIG. 15 is a diagram showing a configuration of a computer system to which a second embodiment of the present invention is applied.
FIG. 16 is a diagram showing an LU area range table used in the second embodiment of the present invention.
FIG. 17 is a flowchart showing a read / write processing procedure according to the second embodiment of the present invention;
FIG. 18 is a diagram showing metadata used in the second embodiment of the present invention.
FIG. 19 is a flowchart showing a procedure of a rearrangement target determination process in the second embodiment of the present invention.
FIG. 20 is a flowchart showing a procedure of rearrangement processing in the second embodiment of the present invention.
FIG. 21 is a diagram showing a configuration of a computer system to which a third embodiment of the present invention is applied.
FIG. 22 is a flowchart showing a processing procedure when a client application reads a file in the third embodiment of the present invention;
FIG. 23 is a flowchart showing a processing procedure when a client application writes a file in the third embodiment of the present invention;
FIG. 24 is a diagram showing a configuration of a computer system to which a fourth embodiment of the present invention is applied.
FIG. 25 is a diagram showing an LU management table 910 according to the fourth embodiment of this invention.
FIG. 26 is a flowchart showing the procedure of a rearrangement target determination process in the fourth embodiment of the present invention.
FIG. 27 is a flowchart showing the procedure of LU construction processing in the fourth embodiment of the present invention.
FIG. 28 is a flowchart showing a procedure of relocation source LU offline processing in the fourth embodiment of the present invention;
FIG. 29 is a diagram showing a configuration of a computer system to which a fifth embodiment of the present invention is applied.
FIG. 30 is a diagram showing an LU area range table 193 according to the fifth embodiment of the present invention.
FIG. 31 is a flowchart showing a processing procedure when an application 840 of a client 800 reads a file in the fifth embodiment of the present invention.
FIG. 32 is a flowchart showing a processing procedure when an application 840 of the client 800 writes a file according to the fifth embodiment of the present invention.
FIG. 33 is a flowchart showing an accessible file response processing procedure according to the fifth embodiment of the present invention;
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 100 ... Host, 110 ... File system, 120, 820 ... OS, 130 ... Manager, 140, 840 ... Application, 160, 260, 460, 860 ... FC interface, 170, 270, 470, 870 ... Network interface, 190 ... Local Disk, 191 ... LU logical location name table, 192 ... LU area range table, 200 ... disk array, 210 ... disk device, 220 ... parity group, 300 ... control unit, 310 ... CPU, 320 ... memory, 321 ... logical / physical Corresponding information, 322 ... Logical volume usage status, 323 ... Copy area management table, 330 ... Cache memory, 400 ... Remote computer, 500 ... Switch, 510 ... Copy control unit, 600 ... Fiber Channel (FC), 700 ... Network, 800 ... Ianto.

Claims (6)

A computer system having a computer and a plurality of storage subsystems connected to the computer,
Each of the plurality of storage subsystems is
Obtaining means for obtaining a usage status of the storage area for each of the storage areas of the storage subsystem;
Moving means for moving data from any of the other storage subsystems according to an instruction from the computer;
Notification means for notifying the computer that the movement has been completed,
The calculator is
A collection unit that collects a usage status of each storage area acquired by the acquisition unit from each of the plurality of storage subsystems;
A file correspondence recording unit that retains a correspondence between a storage area for storing the data and a file managed by a file system executed by the computer;
Using the storage area usage information of each of the plurality of storage subsystems collected by the collection means and the file correspondence recording unit, the usage status of the storage area is collected for each file. A second collection means;
A second calculation for calculating the storage area usage status in each storage subsystem based on the usage status information for each storage area file of each of the plurality of storage subsystems collected by the second collection means Means,
A display unit for displaying at least the calculation result and information on the usage status of the storage area of each storage subsystem collected by the collection unit;
Selection means for receiving an input of selection of each storage area displayed on the display unit from a user of the computer;
A correspondence storage unit that stores a correspondence between a logical position of the data and a storage area in which the data is stored in the storage subsystem;
Updating means for updating the correspondence storage unit when receiving notification that the data movement between the storage subsystems has been completed by the notification means;
The instruction from the computer includes information specifying the storage area selected by the selection means.
A computer system characterized by that. A computer system having a computer and a plurality of storage subsystems connected to the computer,
Each of the plurality of storage subsystems is
Obtaining means for obtaining a usage status of the storage area for each of the storage areas of the storage subsystem;
Moving means for moving data from any of the other storage subsystems according to an instruction from the computer;
Notification means for notifying the computer that the movement has been completed,
The calculator is
A collection unit that collects a usage status of each storage area acquired by the acquisition unit from each of the plurality of storage subsystems;
A file correspondence recording unit that retains a correspondence between a storage area for storing the data and a file managed by a file system executed by the computer;
Using the storage area usage information of each of the plurality of storage subsystems collected by the collection means and the file correspondence recording unit, the usage status of the storage area is collected for each file. A second collection means;
A second calculation for calculating the storage area usage status in each storage subsystem based on the usage status information for each storage area file of each of the plurality of storage subsystems collected by the second collection means Means,
Second selection means for automatically selecting the storage area satisfying a predetermined condition based on the calculated result and storage area information of each storage subsystem collected by the collection means;
A correspondence storage unit that stores a correspondence between a logical position of the data and a storage area in which the data is stored in the storage subsystem;
Updating means for updating the correspondence storage unit when receiving notification that the data movement between the storage subsystems has been completed by the notification means;
The instruction from the computer includes information specifying the storage area selected by the second selection unit.
A computer system characterized by that. The storage area is a computer system according to claim 1 or claim 2 which is a logical storage area, it is characterized in that is formed by using a disk device. The predetermined condition is a condition that the frequency of access to the storage area is a certain value or more.
The computer system according to claim 1 or claim 2, characterized in that. Wherein the computer by the notification unit, when the movement of files between the storage subsystem is informed of the completion, the file correspondence means for updating the recording unit further claim 1 or characterized in that it has a The computer system according to claim 2 . The usage status includes a time when the storage area is occupied when data is read.
The computer system according to any one of claims 1 to 5 , wherein

Images