JP2010257477A - Storage control system and method of controlling the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a storage control subsystem which facilitates managing access to a logical volume from a host. <P>SOLUTION: The storage control system discriminates whether a data pattern of data to be exchanged (data, hereinafter) with a host device suits any of one or more data patterns not for writing included in prepared information on the data pattern not for writing. When negative discrimination results are obtained, the storage control subsystem stores the data in a logical device formed on a disk type storage device. Alternatively, when affirmative discrimination results are obtained, the storage control subsystem eliminates the data without storing it in the logical device. <P>COPYRIGHT: (C)2011,JPO&amp;INPIT

Description

  The present invention relates to a storage control system connected to a host device and a control method for the system.

  For example, in a mission-critical storage system that handles large volumes of data, the data is managed using a storage control subsystem that is configured separately from the host computer (hereinafter simply referred to as “host”). . This storage control subsystem is also called a storage control system, for example, and is a RAID (Redundant Array of Independent Inexpensive Disks) configured by arranging a large number of disk-type storage devices in an array.

  For example, in Japanese Patent Laid-Open No. 11-224164, a disk subsystem having a reduced logical volume and a normal logical volume manages cache memory under the LRU method (see paragraph 5 for the LRU method). It is disclosed that the data of the reduced logical volume stays in the cache memory for a long time by changing the position of the data on the volume and the data on the normal logical volume added to the LRU data string.

JP-A-11-224164.

  By the way, for example, the storage control system includes one or more LDEVs among a plurality of logical storage devices (hereinafter referred to as “LDEV” for short) prepared on one or a plurality of disk type storage devices. Some storage units are provided to a host device as an LU (Logical Unit). In such a storage control system, for example, an LU pair in which one LU is a primary LU and another LU is a secondary LU is formed, and the pre-update data in the primary LU is copied to the secondary LU (that is, A so-called snapshot) may be performed.

  For example, the storage capacities of both LUs constituting the LU pair are the same. For this reason, if the storage capacity of the primary LU is large, an LU having a large storage capacity is required as the secondary LU. Generally, in a storage control system, the same storage capacity as the size of an LU provided to a host device is allocated on a disk type storage device. However, for example, generally, the storage capacity actually used by the host device is only a small part of the storage capacity actually allocated, and most of the storage capacity may be reserved uselessly. When formatting an LU, data having the same size as the LU storage capacity is output from the host terminal, and the data is written to the entire LU.

Accordingly, the present invention achieves at least one of the following objects.
(1) To provide a storage control system that reduces useless free storage areas, prepares a plurality of types of LUs, and stores data in a type of LU selected from the plurality of types of LUs, and a control method for the system .
(2) To provide a storage control system and a control method for the storage capacity consumption of LDEV.

  Other objects of the present invention will become clear from the following description.

  A storage control system according to an aspect of the present invention includes a plurality of channel adapters connected to a host device and receiving data from the host device, and data exchanged between the host devices and connected to the plurality of channel adapters, And one or more memories for storing control information relating to the data; and a plurality of disk adapters connected to the memory and controlling the data to be written to or read from the memory; A plurality of disk drives connected to the plurality of disk adapters, wherein the data is stored under the control of the plurality of disk adapters. The first channel adapter included in the plurality of channel adapters is configured to transfer a plurality of first logical volumes generated based on storage areas of the plurality of first disk drives included in the plurality of disk drives to the host device. And when a data write access is made from the higher-level device to the first logical volume selected from the plurality of first logical volumes, an area reserved in advance in the memory is provided. To write the data. The second channel adapter included in the plurality of channel adapters collectively manages a plurality of second logical volumes generated based on storage areas of the plurality of second disk drives included in the plurality of disk drives, When the information about the third logical volume provided to the host device is received from the host device, if there is a write access of the data from the host device to the third logical volume, the third logical volume The area on the memory required as the second logical volume corresponding to the volume is secured for the area corresponding to the data, and the data is written to the reserved area, corresponding to the third logical volume. The second logical volume to be released is released.

  In the first embodiment of the storage control system, at least one of the one or more memories is a data pattern of non-write target data that should not exist in the second logical volume. The non-write target data pattern is stored. At least one of the second channel adapter and the plurality of disk adapters may be configured such that a data pattern of the data is at least one of the one or more non-write target data patterns stored in the memory. A data pattern comparison is performed to determine whether the data match, and if the result of the comparison is affirmative, the data is discarded.

  In the second embodiment of the storage control system, in the first actual aspect, the non-write target data pattern receives the data from the host device in units of blocks that are data storage units of the second logical volume. This is the format data pattern when

  In the third embodiment of the storage control system, in the first actual aspect, at least one of the second channel adapter and the plurality of disk adapters includes a processor and a data check unit having a register. . The data check circuit unit performs a data pattern comparison as to whether or not a data pattern of the data matches at least one of the one or more non-write target data patterns stored in the memory; The result of the comparison is written into the register. The processor obtains the result of the data pattern comparison from the register, and discards the data if the result of the data pattern comparison is positive.

  In a fourth embodiment of the storage control system, the storage control system further comprises a maintenance terminal for forming a plurality of logical paths connecting the first logical volume and the third logical volume and the higher-level device. Prepare.

  In a fifth embodiment of this storage control system, at least one of the second channel adapter and the plurality of disk adapters is a third logical storage in the third logical volume when the write access is made. The area is associated with the second logical storage area in the second logical volume, and when the result of the data pattern comparison is affirmative, the association is released and the release is performed.

  In a sixth embodiment of the storage control system, at least one of the second channel adapter and the plurality of disk adapters is read from the third logical volume via the second logical volume. When the data is acquired, the data pattern of the data is compared with at least one of the one or more non-write target data patterns stored in the memory, and the data pattern comparison is performed. If the result of the comparison is affirmative, the data existing in the second logical volume is erased.

  In the seventh embodiment of the storage control system, in the sixth actual situation, the second logical volume is composed of a plurality of logical chunks. At least one of the plurality of disk adapters has a data size of the data stored in the logical chunk even if the data pattern comparison result for the data when the write access is made is positive. If the data is smaller than the capacity, the data is stored in the second logical volume, and when data is read, the data is read in units of the logical chunk, and a positive data comparison result is obtained for the read data. If the data is read, the read data is deleted from the logical chunk.

  In an eighth embodiment of the storage control system, when at least one of the second channel adapter and the plurality of disk adapters writes the data to the first logical storage area of the first logical volume, The data in the first logical storage area is read and copied to the second logical storage area of the second logical volume (for example, a snapshot of the data in the first storage area is formed). The data pattern of the read data is compared with at least one of the one or more non-write target data patterns stored in the memory. If the result of the comparison is affirmative, the read data is discarded so as not to exist in the second logical storage area.

  In the ninth embodiment of the storage control system, at least one of the one or more memories is a data pattern of non-write target data that should not exist in the second logical volume. The non-write target data pattern is stored. The second channel adapter is adapted to associate the third logical storage area in the third logical volume with the second logical storage area in the second logical volume, and read access from the higher-level device However, if the access is to a third logical storage area that is not associated with the second logical storage area, at least one of the one or more non-write target data patterns is transmitted to the higher-level device.

  In the tenth embodiment of the storage control system, at least one of the one or more memories stores the attributes of the first and third logical volumes. The second channel adapter reads the attributes of the first and third logical volumes from the memory and notifies the host device.

  A storage control system according to another aspect of the present invention is a storage control system connected to a host device, and includes a physical storage device, a plurality of logical storage devices, a virtual storage device, one or more memories, and a storage And a control unit. The physical storage device physically stores data exchanged between the host device and the storage control system. Each of the plurality of logical storages is a device that is not provided to the host device, and is provided on the physical storage device, and logically stores the physically stored data. The virtual storage device is a device provided to the host device, and has a virtual storage area that is dynamically associated with the logical storage area of the logical storage device. The one or more memories store one or more non-write target data patterns that are data patterns of non-write target data that should not exist in the logical storage device. When the storage control unit receives a read request or a write request from the host device that recognizes the virtual storage unit, the storage control unit includes a virtual storage area in the virtual storage unit and a logical storage area in the logical storage device. Correlation is performed, and data is exchanged between the logical storage area and the host device via the virtual storage area. In addition, the storage control unit performs a data pattern comparison as to whether or not the data pattern of the data is compatible with at least one of the one or more non-write target data patterns stored in the memory. When a data pattern comparison result is obtained, the data is erased so that it does not exist in the logical storage device.

  A storage control system connected to the host device is connected to the host device and receives data from the host device, and is connected to the plurality of channel adapters and exchanges with the host device. One or more memories for storing data and control information related to the data, and a plurality of disk adapters connected to the memory and controlling the data to be written to or read from the memory And a plurality of disk drives connected to the plurality of disk adapters and storing the data under the control of the plurality of disk adapters, the method according to the present invention is a control method of the storage control system. And a storage area of a plurality of first disk drives included in the plurality of disk drives. Providing a plurality of generated first logical volumes to the higher-level device; and writing access of the data from the higher-level device to the first logical volume selected from the plurality of first logical volumes. If there is, a step of writing the data to an area reserved in advance on the memory, and a plurality of second areas generated based on storage areas of a plurality of second disk drives included in the plurality of disk drives When the logical volume is collectively managed and information about the third logical volume provided to the higher-level device is received from the higher-level device, the data is written to the third logical volume from the higher-level device. If accessed, the memory required as the second logical volume corresponding to the third logical volume is used. Reserved area fraction corresponding regions on Li to the data, for the reserved area, and a step of releasing the second logical volume corresponding to the data writing, to the third logical volumes.

  In the first embodiment of the control method, the control method uses a data pattern of non-write target data that should not exist in the second logical volume in at least one of the one or more memories. Storing one or more non-writeable data patterns, and whether the data pattern of the data matches at least one of the one or more non-writeable data patterns stored in the memory A data pattern comparison of whether or not, and if the result of the comparison is affirmative, the method further includes a step of discarding the data. The non-write target data pattern is, for example, a format data pattern when the data is received from the host device in units of blocks that are data storage units of the second logical volume.

  In the second embodiment of the control method, the control method further includes a step of forming a plurality of logical paths connecting the first logical volume and the third logical volume and the higher-level device.

  In the third embodiment of the control method, the control method includes a third logical storage area in the third logical volume and a second logical storage area in the second logical volume when there is a write access. And a step of releasing the association and releasing when the result of the data pattern comparison is affirmative.

  In a fourth embodiment of the control method, when the control method acquires the data read from the third logical volume via the second logical volume, the data pattern of the data is the memory. And a step of performing a data pattern comparison as to whether or not the data pattern matches at least one of the one or more non-write target data patterns stored in the memory, and if the result of the comparison is affirmative, And erasing the data existing in the logical volume.

  In a fifth embodiment of the control method, the control method reads data in the first logical storage area when the data is written to the first logical storage area of the first logical volume, and A step of copying to a second logical storage area of a two logical volume; and a data pattern of the read data is at least one of the one or more non-write target data patterns stored in the memory A step of performing a data pattern comparison as to whether or not conforming, and a step of discarding the read data so as not to exist in the second logical storage area if the result of the comparison is affirmative Have.

  In a sixth embodiment of this control method, the control method uses a data pattern of non-write target data that should not exist in the second logical volume in at least one of the one or more memories. Storing one or more non-write target data patterns, associating a third logical storage area in the third logical volume with a second logical storage area in the second logical volume, and If the read access from the host device is an access to a third logical storage area that is not associated with the second logical storage area, at least one of the one or more non-write target data patterns is transferred to the upper logical unit. And transmitting to the device.

  In a seventh embodiment of the control method, the control method stores the attributes of the first and third logical volumes in at least one of the one or more memories, and the first And further reading the attributes of the first and third logical volumes from the memory and notifying the host device.

According to the present invention, the following effects can be achieved.
(1) A storage control system capable of reducing a useless free storage area, preparing a plurality of types of logical volumes, and storing data in a logical volume selected from the plurality of types of logical volumes, and control of the system A method is provided.
(2) A storage control system and a control method for the system capable of suppressing the consumption of the storage capacity of a logical volume are provided.

1 shows an outline of an external appearance of a storage control system according to an embodiment of the present invention. 1 shows an example of the overall configuration of a storage system according to an embodiment of the present invention. The block diagram which shows the function of the storage control subsystem which concerns on this embodiment. A configuration example of the virtual LDEV 61c, the LDEV pool 68, and the DAT 64 is shown. The structural example of PMT63 is shown. The processing flow of the channel control unit 110C that dynamically associates the virtual LDEV 61c and the LDEV pool 68 is shown. The structural example of the LU-LDEV management table 162b which concerns on this embodiment is shown. An example of a software configuration in the host terminal 200C and an example of a write request output from the host terminal 200C are shown. The structural example of DPT800 is shown. The structural example of 110 C of channel control parts is shown. The processing flow in the storage control subsystem 102 performed when an LU is set is shown. A processing flow of the channel control unit 110C when a write request for the virtual LU 310C is received from the host terminal 110C is shown. A processing flow of the channel control unit 110C when a read request for the virtual LU 310C is received from the host terminal 110C is shown. The figure for demonstrating the 2nd specific example of this embodiment. In the second specific example of this embodiment, a processing flow when a write request is received from the host terminal 200C is shown. The structural example of the disk control part 140A which concerns on the 3rd specific example of this embodiment is shown. In the fourth specific example of this embodiment, a processing flow of the channel control unit 110C when a write request is received from the host terminal 110C is shown. In the fourth specific example of this embodiment, a processing flow of the channel control unit 110C when a read request is received from the host terminal 110C is shown. The structure and process flow which concern on the 5th specific example of this embodiment are shown. The structure and process flow regarding the 6th specific example of this embodiment are shown. The specific processing flow regarding the 6th specific example of this embodiment is shown.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

  First, one concept of this embodiment will be briefly described.

  The storage control system according to the present embodiment includes, for example, one or more write targets in which a data pattern of data exchanged with a host device (hereinafter, data) is included in data pattern information that is not prepared for write It is determined whether or not the external data pattern is met. If a negative determination result is obtained, the storage control subsystem stores the data in a logical device formed on the disk type storage device. On the other hand, if a positive determination result is obtained, the storage control subsystem deletes the data without storing it in the LU.

  Hereinafter, this embodiment will be described in detail.

  FIG. 1 schematically shows the appearance of a storage control system according to an embodiment of the present invention.

  The storage control system 600 can be composed of, for example, a basic chassis 10 and a plurality of additional chassis 12 (may be composed of only the basic chassis 11).

  The basic chassis 10 is a minimum configuration unit of the storage control system 600. The basic housing 10 includes, for example, a plurality of disk storage devices (for example, hard disk drives (HDD)) 300, a plurality of control packages (for example, a channel control unit or a disk control unit described later) 105, and a plurality of power supply units. 400 and a plurality of battery units 500 are detachably provided. The basic housing 10 is provided with a plurality of cooling fans 13.

  Each additional enclosure 12 is an option of the storage control system 600. For example, a maximum of four additional enclosures 12 can be connected to one basic enclosure 10. Each additional enclosure 12 is provided with a plurality of cooling fans 13. Each extension enclosure 12 is provided with a plurality of disk storage devices 300, a plurality of power supply units 400, and a plurality of battery units 500, which are detachably provided. Control is performed by a control function of the control package 105 provided in the housing 10.

  FIG. 2 shows an example of the overall configuration of a storage system according to an embodiment of the present invention.

  The basic components of the storage system 1 are one or a plurality of host terminals 200A to 200D and a storage control system 600.

  Each of the host terminals (higher-level devices) 200A to 200D includes, for example, a CPU (Central Processing Unit), a nonvolatile and / or volatile memory (for example, ROM or RAM), a hard disk, and the like as hardware resources. (For example, a personal computer or a workstation). When the CPUs of the host terminals 200A to 200D read and execute various computer programs stored in the memory, processing in which the computer program and hardware resources (for example, memory) cooperate with each other is performed, and various functions are performed. Is realized. Each of the host terminals 200A to 200D can be connected to the storage control system 600 by various methods.

  For example, the host terminals 200 </ b> A to 200 </ b> B are connected to the storage control system 600 via a first communication network (for example, a LAN, the Internet or a dedicated line, hereinafter referred to as a LAN) 820. Communication between the host terminals 200A and 200B and the storage control system 600 performed via the LAN 820 is performed according to, for example, the TCP / IP protocol. A data access request by specifying a file name (data input / output request in units of files, hereinafter referred to as “file access request”) is transmitted from the host terminals 200A and 200B to the storage control system 600.

  For example, the host terminals 200B and 200C are connected to the storage control system 600 via a second communication network (for example, SAN (Storage Area Network), hereinafter referred to as SAN) 821. Communication between the host terminals 200B and 200C and the storage control system 600 performed via the SAN 821 is performed according to, for example, a fiber channel protocol. For example, a data access request in block units (hereinafter referred to as “block access request”) is transmitted from the host terminals 200B and 200C to the storage control system 600. (A block is a unit of data management in the storage area on the apparatus 300).

  For example, the host terminal 200D is connected to the storage control system 600 without going through a network such as the LAN 820 or the SAN 821. The host terminal 200D can be a mainframe computer, for example. Communication between the host terminal 200D and the storage control system 600 includes, for example, FICON (Fibre Connection: registered trademark), ESCON (Enterprise System Connection: registered trademark), ACONARC (Advanced Connection Architecture: registered trademark), and FIBARC (Fibre Connection). This is performed in accordance with a communication protocol such as Architecture (registered trademark). A block access request is transmitted from the host terminal 200D to the storage control system 600, for example, according to any of these communication protocols.

  For example, a backup storage control system 910 is connected to at least one of the LAN 820 and the SAN 821. The backup storage control system 910 may be a storage control system that stores data in a disk device selected from, for example, one or more disk devices (for example, MO, CD-R, or DVD-RAM). It may be a storage control system such as the storage control system 600 described in detail later, or selected from one or a plurality of tape-based devices (for example, DAT tape, cassette tape, open tape or cartridge tape). A storage control system that stores data in a tape device may also be used. The backup storage control system 910 receives data stored in the storage control system 600 via, for example, the LAN 820 or the SAN 821 (or a host terminal connected to the backup control system 910), and is provided in itself. The data is stored in a storage device (for example, a tape device).

  In addition, for example, a management server 819 may be connected to at least the LAN 820 of the LAN 820 and the SAN 821. For example, the management server 819 may perform communication between the host terminal and another host terminal, or relay communication between the host terminal and the storage control system 600.

  The host terminals 200A to 200D may be connected to each other via a third communication network (for example, a LAN).

  The storage control system 600 is, for example, a RAID ((Redundant Array of Independent Inexpensive Disks)) system. The storage control system 600 performs control according to commands received from the host terminals 200A to 200D. The storage control system 600 includes a storage control subsystem (disk array device) 102 and a maintenance terminal (hereinafter abbreviated as “SVP”) 160. The storage control subsystem 102 includes a storage control device 100 and a storage device unit 101. The storage controller 100 includes one or more channel controllers 110A to 110D, one or more cache memories (hereinafter abbreviated as “CM”) 130, and one or more shared memories (hereinafter referred to as “cache memories”). Shared Memory is abbreviated as “SM”) 120, one or a plurality of disk control units 140 </ b> A to 140 </ b> D, and a connection unit 150. The storage device unit 101 includes one or more physical disk groups 5. Each of the one or more physical disk groups 5 includes a plurality of disk type storage devices 300 arranged in an array.

  Each of the channel controllers 110A to 110D can be configured by a hardware circuit, software, or a combination thereof. Each of the channel control units 110A to 110D is detachable from the storage control device 100 (for example, the basic housing 10), and may be referred to as a channel adapter. Each of the channel control units 110A to 110D includes, for example, a printed circuit board on which a processor, a memory, and the like are mounted, and a control program stored in the memory, and through cooperation between these hardware and software, A predetermined function is realized. Each of the channel control units 110A to 110D is multiplexed (for example, duplexed) so that even if one channel control unit is damaged, it operates with another channel control unit. The channel control units 110A to 110D execute processing according to commands received from the host terminal while referring to control information in the SM 120 (for example, an LU-LDEV management table described later). Of the channel control units 110A to 110D, the channel control unit 110C is taken as an example, and the operations of the disk control units 140A to 140D will be described first. For example, the channel control unit 110C reads from the host terminal 200A or 200B. When an I / O request including the request (input / output request, here, a block access request) is received, a read command is stored in the SM 120 and a cache area is secured in the CM 130. The disk control units 140A to 140D refer to the SM 120 as needed, and when an unprocessed read command is found, data from the disk type storage device 300 (typically, the host terminals 200A to 200D, the disk type storage device 300, User data exchanged between them and read out and stored in the cache area secured in the CM 130. The channel control unit 110C reads the data transferred to the CM 130 from the cache area, and transmits the data to the host terminal 200A or 200B that issued the read request.

  Further, for example, when the channel control unit 110C receives an I / O request including a write request from the host terminal 200B or 200C, the channel control unit 110C stores a write command in the SM 120, secures a cache area in the CM 130, and receives the received I / O request. Data included in the / O request is stored in the secured cache area. Thereafter, the channel control unit 110C reports the completion of writing to the host terminal 200B or 200C that issued the write request. Then, the disk control units 140A to 140D refer to the SM 120 as needed. When an unprocessed write command is found, the disk control unit 140A to 140D reads the data from the cache area secured in the CM 130 according to the write command, It is stored in the disk type storage device 300.

  The above processing can also be performed by the other channel controllers 110A to 110B and 110D. The channel control units 110A to 110B convert the file access request into a block access request (for example, after converting the file name included in the file request into a logical block address according to the file system that the channel control unit 110A to 110B has. ), The above processing is performed.

  The disk controllers 140A to 140D can be configured by hardware circuits, software, or a combination thereof. Each of the channel controllers 140A to 140D can be attached to and detached from the storage controller 100 (for example, the basic chassis 10 or the additional chassis 12), and is sometimes called a disk adapter. Each of the disk controllers 140A to 140D includes, for example, a printed circuit board on which a processor, a memory, and the like are mounted, and a control program stored in the memory. Realize the function. Each of the disk control units 140A to 140D is multiplexed (for example, duplexed) so that even if one disk control unit is damaged, it operates with another disk control unit. The disk control units 140A to 140D perform data communication with each disk type storage device 300 included in each physical disk group 5 while referring to control information in the SM 120 (for example, an LU-LDEV management table described later). It is something to control. Each of the disk control units 140A to 140D and each of the disk storage devices 300 are connected via a communication network such as a SAN, for example, and perform data transfer in units of blocks according to a fiber channel protocol. Further, the disk control units 140A to 140D monitor the state of the disk type storage device 300 as needed, and the monitoring result is transmitted to the SVP 160 via an internal communication network (for example, LAN) 151.

  The one or more CMs 130 are, for example, volatile or nonvolatile memories. The CM 130 secures a cache area, and stores data transmitted and received between the channel controllers 110A to 110D and the disk controllers 140A to 140D. The data may be multiplexed and managed by a plurality of CMs 130.

  One or a plurality of SMs 120 are configured from, for example, a non-volatile memory, and store control information or the like (for example, control information or the like may be multiplexed and managed by a plurality of SMs 120). The control information includes, for example, various commands exchanged between the channel controllers 110A to 110D and the disk controllers 140A to 140D, a cache management table, a disk management table, and an LU-LDEV management table. . The cache management table is, for example, a table in which a correspondence relationship between a cache area and a logical address of an LDEV described later is written. The disk management table is a table for managing each disk type storage device 300. For example, for each disk type storage device 300, the disk ID, the vendor, the storage capacity, the RAID level, the usage status (for example, in use or not used). Used) or the like. The LU-LDEV management table is a table for managing an LDEV described later. For example, for each LDEV, logical path information (for example, port number, target ID and LUN), address management information (for example, a disk type storage device) The correspondence between the physical address on 300 and the logical address in LDEV), storage capacity, and RAID level. The physical address is address information including, for example, the ID of the disk type storage device 300, the disk head number, and the number of sectors. The logical address is address information including, for example, a LUN (Logica Unit Number), an LDEV number, and a logical block address.

  The connection unit 150 connects the channel control units 110A to 110D, the disk control units 140A to 140D, the CM 130, and the SM 120 to each other. Data and commands are exchanged between the channel control units 110A to 110D, the CM 130, the SM 120, and the disk control units 140A to 140D through the connection unit 150. The connection unit 150 includes, for example, a first sub-connection unit through which user data passes and a second connection unit through which control information and the like pass. The channel control units 110A to 110D, the disk control units 140A to 140D, and the CM 130 are connected to the first sub-connection unit, and the channel control units 110A to 110D and the disk control units 140A are connected to the second sub-connection unit. ~ 140D and SM120 are connected. At least the first sub-connection portion of the first sub-connection portion and the second sub-connection portion is, for example, a high-speed bus such as an ultrahigh-speed crossbar switch that performs data transmission by high-speed switching.

  Each of the plurality of disk type storage devices 300 is, for example, a hard disk drive or a semiconductor memory device. A RAID group 2 is configured by a predetermined number of disk storage devices 300 of two or more of the plurality of disk storage devices 300. The RAID group 2 is sometimes called, for example, a parity group or an error collection group, and is a group of the disk type storage devices 300 according to the RAID principle. Two or more disk storage devices 300 belonging to the same RAID group 2 are mounted on different motherboards, for example, and even if one disk storage device 300 fails, the data of the remaining other disk storage devices 300 are used. Thus, the data of the failed disk storage device 300 can be restored. A plurality of LDEVs (Logical Devices) that are logical storage devices are set on the physical storage area provided by the RAID group 2, and one or more LDEVs among the plurality of LDEVs are LUN (Logical Unit Number). ) Is provided from the storage control device 100 to the host terminals 200A to 200D as one LU (Logical Unit) 310. Each LU 310 may be paired, for example, as a primary LU (data copy source LU) with another LU 310 that is a secondary LU (data copy destination LU). In this case, all or a part of the data in the LU 310 ( For example, data before update) may be copied to another LU 310. Further, each LU 310 may be paired with, for example, another LU 310 that is a primary LU as a secondary LU. In this case, all or a part of the data in another LU 310 (for example, data before update) Sometimes copied to the LU 310.

  The SVP 160 is a computer machine for maintaining or managing the storage system 600. The SVP 160 can collect information from each component (for example, each channel control unit 110A to 110D and each disk control unit 140A to 140D) of the storage control system 600 via the communication network 151 such as an internal LAN. it can. Specifically, for example, an OS (operating system), application program, driver software, and the like installed in each component (for example, a channel controller or a disk controller) of the storage control system 600 are generated in the component. The failure occurrence information related to the failure occurrence is output, and the SVP 160 can receive the failure occurrence information. Information received by the SVP 160 includes, for example, a device configuration, a power supply alarm, a temperature alarm, an input / output speed (for example, the number of I / O requests received by the storage control device 100 per fixed time), and the like. Further, for example, the SVP 160 may perform setting of the disk storage device 300, setting of the LDEV, installation of a micro program executed in the channel control units 110A to 110D, and the like in response to an operation by the operator. it can. The SVP 160 can also perform operations such as confirmation of the operation state of the storage control system 600, identification of a failure part, and installation of an operating system executed by the channel control unit 110, for example. Further, for example, the SVP 160 is connected to an external maintenance center (not shown) via a communication network such as a LAN or a telephone line, and the fault received from each component of the storage control system 600 is connected to the external maintenance center. The occurrence information or the like may be notified. In addition, the SVP 160 can be built in the storage control system 600 or can be externally attached.

  The above is the basic description of the storage control system 600. In the storage control system 600, for example, one channel control unit and one disk control unit are integrally configured as one module, and the functions of the channel control unit and the disk control unit are configured by the one module. May be used. For example, the SM 120 and the CM 130 may be integrally configured. Moreover, one LUN may be assigned to each channel control unit, or one LUN may be assigned to a plurality of channel control units. Further, another storage control system may be connected to the storage control system 600. In this case, for example, the primary LU included in the storage control system 600 and the secondary LU included in another storage control system are paired, and the host terminals 200A to 200D connected to the storage control system 600 perform storage control. A secondary LU in another storage control system may be accessed via the system 600. Further, for example, the storage control system 600 may receive both a file access request and a block access request, may receive only a file access request (for example, NAS), or may receive a block access request. You may receive only.

  FIG. 3 is a block diagram showing functions of the storage control subsystem 102 according to the present embodiment. In the following description, for easy understanding, the host terminal 200C is taken as an example among the host terminals 200A to 200D, and a block access request is received from the host terminal 200C among the channel control units 110A to 110D. The channel control unit 110C is taken as an example.

  Between the host terminal 200C and the storage control subsystem 102, one or a plurality of logical communication paths (hereinafter referred to as “logical paths”) 21A to 21D are formed. Each of the logical paths 21A to 21D is formed based on, for example, the number of the port (port to which the host terminal 100A is connected) included in the storage control subsystem 102, the target ID, and the LUN.

  The plurality of LDEVs prepared on one or a plurality of RAID groups 2 include, for example, an LDEV having a normal LDEV attribute (hereinafter referred to as a normal LDEV) 61a and an LDEV having an attribute of a pool LDEV (hereinafter referred to as a pool LDEV). ) 61b. Each LDEV can be changed from the normal LDEV 61a to the pool LDEV 61b by the instruction of the operator of the SVP 160, and conversely, from the pool LDEV 61b to the normal LDEV 61a.

  The normal LDEV 61a is an LDEV accessible by the host terminal 200C. In other words, for example, when the LUN included in the logical path 21A is designated from the host terminal 200C, two normal LDEVs 61a associated with the LUN are provided to the host terminal 200C as the LU 310A. For example, when the LUN included in the logical path 21B is designated from the host terminal 200C, one normal LDEV 61a associated with the LUN is provided to the host terminal 200C as the LU 310B.

  The pool LDEV 61b is a member constituting the LDEV pool 68, and is an LDEV that cannot be accessed by the host. In other words, the pool LDEV 61b is not associated with a LUN that can be designated from the host terminal 200C, and even if a LUN is designated, the pool LDEV 61b itself is not provided to the host terminal 200C.

  One or more virtual LUs 310C provided to the host terminal 200C exist above the LDEV pool 68. The virtual LU 310C is an LU provided to the host terminal 200C. However, unlike the other LUs 310A and 310B, the virtual LU 310C is a virtual LU that does not have a physical data storage area on the RAID group 2. More specifically, the virtual LU 310C is composed of one (or a plurality) of virtual LDEVs 61c. Unlike the normal LDEV 61a and the pool LDEV 61b, the virtual LDEV 61c has a physical data storage area on the RAID group 2. It is a virtual LDEV that does not have. Each of the plurality of virtual addresses included in the virtual LDEV 61c dynamically corresponds to a logical address selected from the plurality of logical addresses included in the pool LDEV 68 via the dynamic addressing table (hereinafter abbreviated as “DAT”) 64. Or the association is released and the logical address is released.

  FIG. 4 shows a configuration example of the virtual LDEV 61c, the LDEV pool 68, and the DAT 64.

  The virtual LDEV 61c is composed of a plurality of virtual chunks 410c, 410c,... Having a certain size (for example, 64 kilobytes), for example. Each virtual chunk 410c is composed of a predetermined number (for example, 128) of logical blocks (for example, one is 512 bytes). Each virtual chunk 410c has a head logical block address (hereinafter abbreviated as “virtual head LBA”), and the virtual chunk 410c can be identified from the virtual head LBA.

  Each LDEV pool 68 is a set of one or more LDEVs 61b, and includes a plurality of logical chunks 410b, 410b,... Having a certain size (for example, 64 kilobytes), for example. Each logical chunk 61b is composed of a predetermined number (for example, 128) of logical blocks (for example, one is 512 bytes). Each logical chunk 410b has a head logical block address (hereinafter abbreviated as “logical head LBA”), and the logical chunk 410c can be specified from the logical head LBA. For example, if the logical head LBA to be accessed is determined, the corresponding chunk number is determined. Specifically, for example, when the logical chunk size is 64 KB (= 128 blocks), the logical chunk number n is an integer part of the logical block number m / 128 (for example, logical blocks with logical block numbers 0 to 127). The logical chunk number having the logical block number is zero, and the logical chunk number having the logical block number of 128 to 255 is 1).

  The DAT 64 is registered in a storage area (eg, SM 120) in the storage control subsystem 102 from an external terminal (eg, SVP 160). The DAT 64 is prepared for each virtual LDEV 61c, and is a table for associating each virtual chunk 410c of the virtual LDEV 61c with each logical chunk 410b in each of the one or more LDEV pools 68. Specifically, for example, the DAT 64 includes, for each virtual head LBA, the identification information (for example, LDEV pool number) of the LDEV pool 68 associated with the virtual head LBA, and the logical head LBA in the LDEV pool 68. Is described. The description content of the DAT 64 is updated by the channel control unit 110A at a predetermined timing, for example, when the channel control unit 110A receives an I / O request. The channel control unit 110 </ b> A refers to a pool management table (hereinafter abbreviated as “PMT”) 63 to update the description content of the DAT 64.

  FIG. 5 shows a configuration example of the PMT 63.

  The PMT 63 exists for each LDEV pool 68, and is registered in a storage area (for example, SM 120) in the storage control subsystem 102 from an external terminal (for example, the SVP 160). An entry number is assigned to each logical chunk 401b of the LDEV pool 68. In each PMT 63, the number of the first empty entry in the corresponding LDEV pool 68 (in other words, the chunk number having the smallest logical head LBA) is written in the queue format in the queue format.

  Hereinafter, the processing flow of the channel control unit 110C that dynamically associates the virtual LDEV 61c and the LDEV pool 68 will be described with reference to FIG.

  The host terminal 200C outputs an I / O request for the virtual LU 310C in the same manner as when accessing the normal LU 310A. For example, when the channel control unit 110C receives the I / O request from the host terminal 200C (step S11), the channel control unit 110C determines whether the virtual head LBA calculated from the I / O request and the logical head LBA of the LDEV pool 68 are Perform the association. Specifically, for example, the channel control unit 110C refers to the DAT 64 corresponding to the virtual LDEV 61c included in the virtual LU 310C that has received the I / O request, and acquires the LDEV pool number corresponding to the calculated virtual head LBA. (S12). Next, the channel control unit 110C acquires the logical head LBA corresponding to the first empty entry number written in the PMT 63 from the PMT 63 corresponding to the acquired LDEV pool number, and uses the logical head LBA as the virtual head LBA. Registration is made at a location corresponding to the first LBA (location on DAT 64) (S13). As a result, the logical head LBA is associated with the virtual head LBA, and data can be read from or written to the logical chunk having the logical head LBA.

  For example, the channel control unit 110C can release the logical head LBA written in the DAT 64 after the processing based on the I / O request is completed (Y in S14). Specifically, for example, the channel control unit 110C deletes the logical head LBA registered in the DAT 64 from the DAT 64, and stores the logical head LBA and the corresponding entry number in the PMT 64 in the PMT 64 (S15).

  Through the flow as described above, the dynamic association between the virtual LU 310C (virtual LDEV 61c) and the LDEV pool 68 and the cancellation thereof are performed. As described above, an LDEV pool 68 composed of pool LDEVs 61b, 61b,... To which access from which host terminal is not assigned in advance is prepared, and a storage area in the LDEV pool 68 is flexibly allocated to a virtual LU 310C (virtual LDEV 61c). By associating, a free storage area can be used effectively.

  FIG. 7 shows a configuration example of the LU-LDEV management table 162b according to this embodiment.

  The LU-LDEV management table 162b is registered in a storage area (for example, SM 120) in the storage control subsystem 102. In the LU-LDEV management table 162b, for example, for each port (port to which a host terminal is connected) included in the storage control system 600, a port number, one or more target IDs belonging to the port number, and each target ID LDEV information relating to one or more LUNs belonging to and one or more LDEVs belonging to each LUN are recorded. The LDEV information of each LDEV includes, for example, an LDEV number, storage capacity, RAID level, the above-described address management information, LDEV attribute, DAT-ID, LDEV pool number, and status.

  Examples of the LDEV attribute include the above-described pool LDEV, normal LDEV, and virtual LDEV.

  Since the LDEV having the LDEV attribute “pool LDEV” (the aforementioned pool LDEV 61b) is not accessed by the host terminal, logical path information (for example, LUN) is not associated with the pool LDEV 61b. Further, an LDEV pool number is associated with the pool LDEV 61b so as to identify which LDEV pool 68 is a member of.

  The LDEV having the LDEV attribute “virtual LDEV” (the above-described virtual LDEV 61c) is associated with the ID (for example, zero) of the DAT 64 corresponding to the LDEV 61c. By referring to this DAT-ID, it is possible to know which DAT should be referred to when which virtual LDEV 61c (virtual LU 310C) is accessed. Further, the LDEV pool number is associated with the virtual LDEV 61c so as to identify which LDEV pool 68 is a member of.

  The DEV-ID and the LDEV pool number are not associated with the LDEV having the LDEV attribute “normal LDEV” (the above-described normal LDEV 61a).

  The LDEV status includes, for example, “Ready” (for example, mounting state) indicating an accessible state and “Not Ready” (for example, unmounted state) indicating an inaccessible state. .

  By the way, the channel control unit 110C sends the host data to the disk control unit 140A via the CM 130 based on the data pattern of data received from the host terminal 200C (or data acquired from the LU, hereinafter referred to as “host data”). To 140D, or a process for determining whether to erase the host data without passing it to the disk control units 140A to 140D (in other words, storing the host data in the disk-type storage device 300, or host data To determine whether to delete the data without storing it in the disk-type storage device 300. This process is useful, for example, when format data described later is output from the host terminal 200C in order to format an LU (one or more LDEVs). Details will be described below.

  FIG. 8 shows a software configuration example in the host terminal 200C and an example of a write request output from the host terminal 200C.

  A control unit (for example, CPU) 323, a memory 321, a storage device (for example, a hard disk drive) 322, and a user interface (for example, an input device such as a keyboard) 474 are connected to the internal bus 324 of the host terminal 200C. An operating system (hereinafter referred to as “OS”) 472 such as UNIX (registered trademark) 472 and an application program 471 operating on the OS 472 are loaded from the storage device 322 and stored in the memory 321 of the host terminal 200C. The The OS 472 and the application program 471 function by being read by the control unit 323.

  The user interface 474 notifies the application program 471 or the OS 472 of the result of the user operation.

  For example, when the application program 471 receives an instruction to write a data file to a user-specified LU from the user interface, the application program 471 issues a file write instruction to the OS 472 to write the data file to the user-specified LU. Send.

  For example, when receiving a file write instruction from the application program 471, the OS 472 refers to the file system 475 that the OS 472 owns, converts the data file into block data 476, and converts the block data 476 into a user-specified LU. A first write request (block access request, for example, a SCSI command) meaning writing is generated, and the first write request is transmitted to the storage control subsystem 102.

  For example, when a predetermined event occurs (for example, when the power is turned on or when the OS 472 is restarted), the OS 472 requests the storage control subsystem 102 for system information related to the storage control subsystem 102. A system information request command (for example, a read capacity command or a device discovery command) is transmitted to the storage control subsystem 102. In this case, the host terminal 200C receives system information (for example, information (for example, logical path information and storage capacity) regarding all LUs accessible by the host terminal 200C) from the storage control subsystem 102, and receives the system information. The data is stored in a storage device (for example, a memory or a hard disk) in the host terminal 200C.

  For example, when the OS 472 receives a predetermined command from the user interface 474 without going through the application program 471, the OS 472 generates a second write request having host data having a certain data pattern based on the command. To the storage control subsystem 102. Specifically, for example, when the OS 472 receives a format instruction meaning that the user-specified LU is formatted from the user interface 474, the OS 472 includes format data 473 having the same size as the storage capacity of the user-specified LU. A second write request (block access request, for example, a SCSI write command) is generated, and the second write request is transmitted to the storage control subsystem 102.

  In other words, in the present embodiment, host data with a non-constant data pattern (that is, block data with no regularity of data) 476 is included in the write request received from the host terminal 300C by the channel control unit 110C of the storage control subsystem 102. And a second write request having host data having a constant data pattern (that is, block data having a regularity of data, for example, format data 473). If the write request received by the channel control unit 110C is the second write request, and the format data 473 is included in the second write request, the format data 473 is stored in the LU specified by the user. Since the size is the same as the capacity, the format data 473 is written in the entire user-specified LU unless the device described below is applied to one of the feature points of this embodiment. For this reason, since user data (for example, document data created by the user) is not stored, data is stored in a logical storage area (LDEV storage area) that is originally free. This disadvantage is greater when the user-specified write destination LU is a virtual LU 310C than when it is a normal LU 310A, 310B. This is because the free storage area (eg, free chunk) of the normal LDEV 61a included in the normal LUs 310A and 310B is allocated to a host terminal predetermined by a logical path and is not dynamically allocated, but the virtual LU 310C is dynamically allocated. This is because the free storage area (for example, free chunks) of the LDEV pool 68 associated with is not assigned to a predetermined host terminal but is dynamically assigned. In other words, if the free storage area of the LDEV pool 68 is unnecessarily reduced, the merit of providing the virtual LU 310C and performing the above-described dynamic association becomes difficult to appear.

  Therefore, in the present embodiment, the device described below is provided. Hereinafter, the basic concept of the main part in this embodiment and a specific example will be described separately.

  (1) Basic concept of this embodiment.

  As shown in FIG. 8, the memory 520 (for example, SM 120) included in the storage control subsystem 102 has data (hereinafter referred to as “data not to be written”) that should not exist in the LU 310 (one or more LDEVs). Data pattern information 530 (for example, a data pattern table (hereinafter abbreviated as “DPT”) 800 described later) indicating one or more data patterns is stored.

  The storage control subsystem 102 includes a data storage control unit 481 (for example, channel control units 110C and 110D having the data check circuit 67 or disk control units 140A to 140C). The data storage control unit 481 determines whether the data pattern of data exchanged with the host terminal 200C (hereinafter referred to as host data) matches either one or a plurality of data patterns included in the data pattern information (for example, Whether or not they match) is determined. When a negative determination result is obtained, the data storage control unit 481 stores the host data in the LU 310 or transfers the host data to the host terminal 200C. On the other hand, if a positive determination result is obtained, the data storage control unit 481 does not store the host data in the LU because the host data is data not to be written (if it exists in the LU, it is deleted. ).

  According to this basic concept, for example, when the storage control subsystem 102 receives a first write request, the data pattern of the block data 476 included in the first write request is not constant. And the block data 476 is stored in the LU 310. However, when the storage control subsystem 102 receives the second write request, the data pattern of the block data (for example, format data 473) included in the second write request is constant and therefore included in the data pattern information 530. The block data is not stored in the LU 310 (if it exists in the LU 310, it is erased).

  (2) First specific example of this embodiment.

  The first specific example is an example in which the channel control unit 110C receives an I / O request for the virtual LU 310C (virtual LDEV 61c) from the host terminal 300C. Details will be described below.

  FIG. 9 shows a configuration example of the DPT 800 shown in FIG.

  For example, the DPT 800 is registered in at least one of a memory mounted in the channel control unit 110C, a memory mounted in the disk control units 140A to 140D, the CM 130, and the SM 120. The DPT 800 is registered from, for example, an external terminal (for example, SVP 160). In the DPT 800, for each LDEV (for example, virtual LDEV 61c), one or a plurality of data patterns and a valid flag indicating whether or not the data pattern registered in the LDEV is valid are registered.

  Each of the one or more data patterns has a certain pattern, for example, a data pattern of format data. Since the data pattern of the format data is different for each OS (for example, type or version), for example, data stored in which LDEV is input based on what OS is output. The registration of one or more data patterns in the DPT 800 may be performed manually by a user of an external terminal or automatically by a computer program. Further, at least one of the data length and the maximum data length of the data pattern is determined in advance. For example, at least one of the data length and the maximum data length of the data pattern is a multiple or a divisor of the storage size of the logical chunk.

  For example, if the valid flag of each LDEV is “1”, it indicates that the data pattern associated with the LDEV is valid, and if it is “0”, for example, it indicates that it is invalid. If valid, the data pattern is used to compare with the data pattern of the host data, and if invalid (for example, if no data pattern is registered in the corresponding entry), such comparison is performed. No processing is performed.

  The LDEV information registered in the DPT 800 may be virtual LDEV 61c information or normal LDEV 61a information, but the virtual LDEV 61c information is considered to be more desirable. In the DPT 800, a data pattern may be registered for each OS.

  FIG. 10 shows a configuration example of the channel control unit 110C.

  The channel control unit 110C is configured by a board in which hardware elements are integrated into a unit. The channel control unit 110 </ b> C includes a board connection connector (not shown), and the channel control unit 110 </ b> C is electrically connected to the storage control device 100 by fitting the board connection connector to a predetermined connector of the storage control device 100. Connected. The channel control unit 110C includes, for example, a host interface unit (hereinafter referred to as “host I / F”) 711, an SVP interface unit (hereinafter referred to as “SVPI / F”) 51, a host-side memory 113, one or more An input / output control unit 771, a CM data transfer circuit 710, an SM data transfer circuit 740, an internal bus 750, and a data check circuit 67 are provided. A CHP (channel processor) 119, a host-side memory 113, a CM data transfer circuit 710, and an SM data transfer circuit 740, which will be described later, are connected to the internal bus 750.

  The host I / F 711 is a communication interface for performing communication with the host terminal 200C, and receives, for example, a block access request transmitted from the host terminal 200C according to the fiber channel protocol.

  The SVPI / F 51 is connected to the SVP 160 via a communication network such as the internal LAN 150 and is connected to a CHP 119 described later. The SVPI / F 51 is a communication interface (for example, a LAN controller) for controlling communication between the SVP 160 and the CHP 119.

  The host-side memory 113 stores various programs and data used by the CHP 119, for example. Specifically, for example, data such as a data pattern used by the data check circuit 67 is stored at a predetermined address in the host-side memory 113. This data pattern or the like may be, for example, the DPT 800 itself read from the SM 120, or a data pattern for an LDEV in which data output from the channel control unit 110C can be stored.

  The data check circuit 67 is provided at a predetermined location of the channel control unit 110C, for example, between the CM data transfer circuit 710 and the CM 130. The data check circuit 67 can be configured by hardware, software, or a combination thereof (for example, a pure hardware circuit). The data check circuit 67 compares the data pattern of the host data with one or more data patterns (for example, DPT 800) stored in the host-side memory 1103 (hereinafter referred to as “data pattern comparison”). The data check circuit 67 has a register (not shown), and the register has a result of data pattern comparison (that is, whether one or a plurality of data patterns conforms to the data pattern of the host data). Or not).

  The CM data transfer circuit 710 is hardware that executes data transfer between the host terminal 110 </ b> C and the CM 130. The CM data transfer circuit 710 transfers the data from the CM 130 to the host terminal 110C according to the instruction of the CHP 119, or conversely transfers the data from the host terminal 110C to the CM 130.

  The SM data transfer circuit 740 is hardware for executing data transfer between the host side memory 113 and the SM 120. The SM data transfer circuit 740 transfers data from the SM 120 to the host side memory 113 in accordance with an instruction from the CHP 119, and conversely transfers data from the host side memory 113 to the SM 120.

  Each input / output control unit 771 exchanges data and commands with the disk control units 140A to 140D, the CM 130, the SM 120, and the SVP 160 (for example, a microprocessor unit). Each input / output control unit 771 includes a CHP (channel processor) 121 and a CHP memory 119.

  The CHP (channel processor) 121 is, for example, a microprocessor, and controls the entire channel control unit 110C, and controls communication with the disk control units 140A to 140D, the host terminal 200C, and the SVP 160. The CHP 121 functions as the channel control unit 110C by executing various computer programs stored in the CHP memory 121 (or the host-side memory 113). For example, the CHP 121 detects a result of the data pattern comparison by reading a register (not shown) of the data check circuit 67 and determines whether to erase the host data according to the result of the comparison. .

  The CHP memory 119 is a volatile or nonvolatile memory (for example, NVRAM (Non Volatile RAM)) and stores, for example, a computer program that controls the CHP 121.

  The above is the configuration example of the channel control unit 110C. The configuration shown in FIG. 10 is merely an example, and a configuration other than the above configuration can also be employed. The configuration can also be applied to another channel control unit 110D that receives a block access request.

  FIG. 11 shows a processing flow in the storage control subsystem 102 performed when an LU is set.

  The process shown in this figure is executed by, for example, the channel control unit 110C.

  The channel control unit 110C receives from the SVP 160 a selection as to whether to define a virtual LU or a normal LU (step S290).

  In S290, when a virtual LU is selected (YES in S290), the channel control unit 110C receives an LDEV pool creation request from the SVP 160 (or an external terminal connected thereto). In response to the request, the channel control unit 110C provides the LDEV pool edit screen 552 to the SVP 160. This screen 552 includes, for example, an input field for an LDEV pool number to be created and an LDEV number input field for adding, removing, or editing the LDEV pool. Further, for example, the screen 552 is provided with a tool for accepting selection of whether to add, remove, or edit the LDEV number input in the input of the LDEV number.

  The channel control unit 110C creates (or edits) the LDEV pool according to the information input on the LDEV pool edit screen 552 (in other words, updates the LU-LDEV management table 162b) (step S300).

  Thereafter, the channel control unit 110C provides the LU setting screen 551 to the SVP 160. The screen 551 includes input fields for a port number, target ID, LUN, LDEV number, and storage capacity associated with the virtual LU (virtual LDEV).

  The channel control unit 110C defines a virtual LU according to the information input on the LU setting screen 551 (in other words, information on a new virtual LDEV having the LDEV attribute “virtual LDEV” is stored in the LU-LDEV management table). 162b) (S310).

  After S310, the channel control unit 110C receives a write target for receiving an input of a data pattern (that is, a data pattern of non-write target data) associated with a virtual LDEV (virtual LU) registered using the virtual LU setting screen 551. An external data pattern registration screen 553 is provided to the SVP 160. When one or more data patterns are input to this screen 553, the channel control unit 110C converts the input one or more data patterns into information (for example, LDEV number) input to the virtual LU setting screen 551. Correspondingly, it is registered in the DPT 800 (S320).

  After that, the channel control unit 110C receives designation of which LDEV pool the virtual LU defined in S310 uses (for example, receives input of the number of the LDEV pool), and designates the designated LDEV pool as a virtual LU (virtual LU). (In other words, the input LDEV pool number is registered in the LU-LDEV management table 162b) (S320).

  In S290, when a virtual LU is selected (YES in S290), the channel control unit 110C provides the above-described LU setting screen 551 to the SVP 160. The channel control unit 110C defines the LU according to the information input on the LU setting screen 551 (in this case, information on one or more new normal LDEVs having an LDEV attribute of “normal LDEV” is managed by LU-LDEV). Registered in the table 162b) (S291).

  With the above flow, the definition of the virtual LU (virtual LDEV) or the normal LU (one or more normal LDEVs) is completed. With this processing flow, the host terminal 300C can issue an I / O request to either the virtual LU or the normal LU.

  Note that the channel control unit 110C receives at least one piece of information (for example, virtual storage capacity in the case of a virtual LDEV) from among a plurality of types of information input in this processing flow, as a predetermined command (for example, In response to the SCSI command (Inquiry command or read capacity command), the host terminal 200C is notified. Further, for example, when the channel control unit 110C receives an I / O request for the virtual LU 310C from the host terminal 200C, if the virtual LU 310C (virtual LDEV 61c) is not associated with at least one LDEV pool 68, the I / O request is made. / O Reject the request. In this case, for example, the channel control unit 110C may notify the host terminal 200C of that fact, or may cause the host terminal 200C to time out without returning any response.

  FIG. 12 shows a processing flow of the channel control unit 110C when a write request for the virtual LU 310C is received from the host terminal 110C.

  In the write request received from the host terminal 200C by the channel control unit 110C, in addition to the host data, information indicating in which storage area the host data is written, specifically, for example, which block of which LU (for example, LUN) The number of blocks (for example, the number of blocks) to be written from (for example, the top logical block address) is included. In this embodiment, one block which is a data storage unit in each LDEV is, for example, 512 bytes.

  When the channel control unit 110C receives a write request for the virtual LU 310C from the host terminal 200C, the CHP 119 of the channel control unit 110C specifies from the host terminal 200C with reference to the DAT 64 associated with the virtual LU 310 (virtual LDEV 61c). It is determined whether the virtual storage area (for example, the virtual head LBA) that has been assigned is associated with the logical storage area (for example, the logical head LBA) of the LDEV pool 68 (S510).

  If a positive result is obtained as a result of the determination in S510 (YES in S510), the CHP 119 executes a process of S540 described later.

  On the other hand, if a negative result is obtained as a result of the determination in S510 (NO in S510), the CHP 119 performs the process described with reference to FIG. 6 to write from the free logical storage area in the LDEV pool 68. A logical storage area having the same size as the virtual storage area included in the request (a logical storage area that can store host data) is secured, and the logical storage area is associated with the virtual storage area (S520). Further, the CHP 119 secures a cache area on the CM 130 having the same size as the secured logical storage area (S530).

  After YES in S510 or after S530, the CHP 119 reads and reads all or a part of the data pattern written in the DPT 800 (for example, only the data pattern corresponding to the virtual LDEV 61c specified in the write request). The data pattern is set in the data check circuit 67 (for example, the register of the circuit 67), and the data check circuit 67 is activated (S540). Next, the CHP 119 stores the host data included in the received write request in the cache area secured in S530 via the CM data transfer circuit 710 and the data check circuit 67 (S550).

  In S550, the data check circuit 67 compares whether the host data input from the CM data transfer circuit 710 conforms to the data pattern set in S540. The data check circuit 67 writes the data pattern comparison result to the register.

  Here, for example, if the host data has a certain pattern such as format data, the data pattern comparison result is positive, while if the host data does not have a certain pattern, The data pattern comparison result is negative.

  The CHP 119 accesses the register of the data check circuit 67, and when a positive data comparison result is written in the register, the logical storage area (the storage area in the LDEV pool 68) allocated to the virtual storage area in S520. ) Is released (that is, the logical head LBA associated with DAT64 is removed) (S570). Further, the CHP 119 releases the cache area secured in S530 (S580). As a result, the host data written in the cache area is deleted and never written into the LDEV pool 68.

  On the other hand, if a negative data result is written in the register of the data check circuit 67, the CHP 119 ends this processing flow. As a result, the host data written in the cache area is acquired by any of the disk controllers 140A to 140D and stored in the logical storage area allocated in S520 (that is, the pool LDEV 61b in the LDEV pool).

  FIG. 13 shows a processing flow of the channel control unit 110C when a read request for the virtual LU 310C is received from the host terminal 110C.

  In the read request received from the host terminal 200C by the channel control unit 110C, information on which storage area data is read from, specifically, for example, which block (for example, the first logical block address) of which LU (for example, LUN) The number of blocks (for example, the number of blocks) of data to be read is included.

  When the channel control unit 110C receives a read request for the virtual LU 310C from the host terminal 200C, the CHP 119 of the channel control unit 110C designates from the host terminal 200C with reference to the DAT 64 associated with the virtual LU 310 (virtual LDEV 61c). It is determined whether or not the virtual storage area (for example, the virtual head LBA) is associated with the logical storage area (for example, the logical head LBA) of the LDEV pool 68 (S601).

  If a positive result is obtained as a result of the determination in S601 (YES in S510), the CHP 119 performs the process described with reference to FIG. 6 if the cache area is not secured (NO in S602). A logical storage area having the same size as the virtual storage area included in the read request is secured from the free logical storage area in the LDEV pool 68, and the logical storage area is associated with the virtual storage area (S603). Also, the CHP 119 reserves a cache area on the CM 130 having the same size as the secured logical storage area (S604).

  In S602, if the cache area has been secured (YES in S602), the CHP 119 performs the following S605.

  After YES in S602 or after S604, the CHP 119 reads and reads all or a part of the data pattern written in the DPT 800 (for example, only the data pattern corresponding to the virtual LDEV 61c specified in the read request). The data pattern is set in the data check circuit 67 and the data check circuit 67 is activated (S605). Next, the CHP 119 reads data from the secured cache area and transfers the data to the host terminal 200C (S606).

  In S606, the data check circuit 67 compares whether the data output from the CM data transfer circuit 710 conforms to the data pattern set in S605. The data check circuit 67 writes the data pattern comparison result to the register.

  The CHP 119 accesses the register of the data check circuit 67, and when a positive data comparison result is written in the register, the data transferred in S606 is erased from the LDEV pool and assigned to the virtual storage area. The allocated logical storage area is released (S608). By this S608, the non-write target data that existed in the LDEV pool 68 for some reason is erased from the LDEV pool 68.

  Thereafter, the CHP 119 releases the secured cache area (S609).

  When a negative result is obtained as a result of the determination in S601 (YES in S601), that is, when the CHP 119 receives a read request for a virtual storage area to which a logical storage area is not allocated, it is written in the DPT 800. All or a part of the data pattern (for example, only the data pattern corresponding to the virtual LDEV 61c specified by the read request) is read, and the read data pattern is notified to the host terminal 200C (S610).

  As described above, according to the first specific example described above, when the channel control unit 110C receives a write request from the host terminal 200C, the data pattern of the host data included in the write request is registered in advance 1 Or, it is compared whether or not it matches one of a plurality of data patterns (data patterns of data not to be written), and if the comparison result is affirmative, the host data is deleted without being stored in the LDEV. When the channel control unit 110C receives a read request from the host terminal 200C, the data pattern of data read from the LDEV (data transferred to the host terminal 200C) is registered in advance as one or a plurality of data It is compared whether or not it matches any of the patterns (data patterns of data not to be written), and if the comparison result is affirmative, the read data is erased from the LDEV. As a result, more free storage areas can be created in the LDEV.

  (2) Second specific example of the present embodiment.

  FIG. 14 is a diagram for explaining a second specific example. FIG. 15 shows a processing flow when a write request is received from the host terminal 200C in the second specific example. Hereinafter, a second specific example will be described with reference to both the drawings.

  For example, a unit for allocating a storage area in the disk type storage device 300 is a logical chunk constituting an LDEV. The storage capacity of logical chunks (and virtual chunks) is M × N bytes (M is a natural number) when one block in the LDEV is N bytes (N is a natural number). For example, when N = 512 and M = 128, the storage capacity of the logical chunk is 64 kilobytes (KB).

  In this second specific example, the data size of the host data received from the host is a divisor of the storage capacity of the logical chunk (that is, less than the storage capacity), for example, a quarter of the storage capacity of the logical chunk. To do. Specifically, for example, when the storage capacity of the logical chunk 410b is 64 KB16, it is assumed that 16 KB of host data 610 is received. It is assumed that the data size of the data pattern of the non-write target data registered in the DPT 800 is less than the storage capacity of the logical chunk, for example, 512B.

  Even in this case, the data check circuit 67 of the channel control unit 110C performs the processing of S510 to S550 in FIG. 12, and compares the 16 KB host data 610 with the 512B non-write target data pattern. In this case, the data size of the host data 610 is less than the storage capacity of the logical chunk, but if all 512B parts constituting the data 601 match the non-write target data pattern, the comparison result of the data pattern comparison is positive. A typical result is written in the register 602 of the data check circuit 67.

  When the CHP 119 refers to the register 602 and obtains a positive comparison result (that is, the fact that the host data 610 matches the non-write target data pattern) (YES in S560 of FIG. 15), the CHP 119 It is determined whether the data size is less than the storage capacity of the logical chunk (S560A).

  If a positive determination result is obtained as a result of the determination in S560A (YES in S560A), the CHP 119 ends this processing flow, whereas if a negative determination result is obtained (NO in S560), The processing of S570 to S580 is performed.

  Thus, when the data size of the host data is less than the storage capacity of the logical chunk, even if the data pattern comparison result is affirmative, the data is written in the logical chunk 410b. If this is repeated, as a result of the host data being written to the full logical chunk 410b, a set of host data having a certain data pattern exists in the entire logical chunk 410b. Specifically, for example, when four write requests each having four 16 KB host data are received, even if a positive data comparison result is obtained for all host data, all host data (that is, data pattern conformance) is obtained. Host data) is written into the logical chunk 410b, and the logical chunk 410 having a storage capacity of 64 KB is filled with four 16 KB host data.

  As shown in FIG. 14, when the host data is read from the LDEV, the host data is read in units of logical chunks. For this reason, as described above, when a plurality of data pattern compatible host data exist in one logical chunk, a data group including a plurality of data pattern compatible host data is read as one host data.

  The data check circuit 67 also performs data pattern comparison on the read host data and writes the comparison result in the register 602. When the CHP 119 acquires a positive data pattern comparison result from the register 602 for the host data read from the logical chunk 410b, the CHP 119 deletes the host data in the logical chunk 410b and makes the logical chunk 410b free logical. Let it be a chunk 410b. As a result, the logical chunk 410b existing in the LDEV pool 68 becomes a free logical chunk 410b and can be dynamically associated with the virtual chunk 410c.

  The trigger for reading the host data in units of logical chunks includes, for example, when receiving a read request from the host terminal 200C, creating a snapshot in the logical chunk specified by the write request from the host terminal 200C, In some cases, parity data is generated using host data in a logical chunk.

  According to this second specific example, even if host data having a constant data pattern exists in a logical chunk, it can be eliminated.

  (3) A third specific example of the present embodiment.

  In the third specific example, the data check circuit 67 is in the disk control unit.

  FIG. 16 shows a configuration example of the disk control unit 140A according to the third specific example. Since the disk controllers 140A to 140D are substantially the same in configuration, the disk controller 140A will be described as a representative example.

  The disk control unit 140A is formed as a unitized board (for example, a disk adapter). The disk control unit 140A includes a board connection connector (not shown). The disk control unit 140A is electrically connected to the storage control device 100 by fitting the board connection connector to a predetermined connector of the storage control device 100. Connected to. The disk control unit 140A includes a disk interface unit (hereinafter abbreviated as “disk I / F”) 141, a disk-side memory 115, one or a plurality of input / output control units 591, and an SVP interface unit (hereinafter “SVPI / F”). 50), a CM data transfer circuit 713, and an SM data transfer circuit 741 are mounted.

  The disk I / F 711 is a communication interface for performing communication with the disk storage device 300, and is connected to the disk storage device 300 via a communication network such as a SAN, for example.

  Each of the one or more input / output control units 591 controls processing performed based on the I / O request received from the host terminals 200A to 200D, and is, for example, a microprocessor unit. For example, the DKP 123 and the DKP memory 122 are mounted on the input / output control unit 591.

  The DKP 123 is, for example, a microprocessor, and controls the entire disk control unit 140, and controls communication with the channel control units 110A to 110D, the disk storage device 300, and the SVP 160. The DKP 123 functions as the disk control unit 140A by executing various computer programs stored in the DKP memory 122 (or the disk-side memory 115). The functions exhibited by the disk control unit 140A include, for example, control of data storage and reading to the disk type storage device 300 based on the RAID level, and control of copying and backup of data stored in the disk type storage device 300. .

  The DKP memory 122 is a volatile or non-volatile memory (for example, NVRAM (Non Volatile RAM)), and stores, for example, a computer program that controls the DKP 123. The contents of the program stored in the DKP memory 122 can be written or rewritten by an instruction from the SVP 160 or a NAS manager 806 described later, for example.

  The SVPI / F 50 is connected to the SVP 160 via a communication network such as the internal LAN 150 and is connected to the DKP 123. The SVPI / F 50 is a communication interface (for example, a LAN controller) for controlling communication between the SVP 160 and the DKP 123.

  For example, the CM data transfer circuit 713 receives user data from the CM 130 and transfers it to the disk-side memory 115 under the control of the DKP 123, or transfers user data acquired from the disk-type storage device 300 to the CM 713. To do.

  For example, under the control of the DKP 123, the SM data transfer circuit 741 transfers control information (for example, an interprocessor message) from the DKP 123 to the CHP (channel processor) to the SM 120, or transfers control information received from the SM 120 to the DKP 123. To do.

  The disk-side memory 115 is a volatile or non-volatile memory, for example, temporarily storing user data, or storing one or more non-write target data patterns read by the DKP 123.

  In the third specific example, the data check circuit 67 described above is provided between the CM 130 and the CM data transfer circuit 713, but is not limited thereto, and may be provided at another location on the disk control unit 140A.

  In the third specific example, the DKP 123 performs the same process as the CHP 119 in the process based on the comparison result between the host data data pattern and the non-write target data pattern.

  For example, the data pattern of the host data output by the data check circuit 67 to the disk type storage device 300 (or read from the disk type storage device 300) is the non-write target data pattern registered in the disk side memory 115. Are compared, and the comparison result is written in a register (not shown). When the DKP 123 acquires a positive data pattern comparison result from the register, the DKP 123 deletes the host data.

  If the data check circuit 67 is mounted on the disk control unit 140A as in the third specific example, the following processing can be further executed.

  (A) When collection read is performed.

  Collection read refers to data from a plurality of logical chunks respectively possessed by a plurality of other disk storage devices 300 in the same RAID group 2 when the disk control unit becomes inaccessible to one disk storage device 300. Is to read. In this case, the disk control unit 140A can perform exclusive OR of a plurality of data respectively read from the plurality of logical chunks and restore the data. When data is read from a logical chunk in this collection read, a logical storage area release process of the LDEV pool is executed.

  According to the third specific example, the disk controllers 140A to 140D perform comparison of data patterns and processing based on the comparison result.

  (B) When drive copy is performed.

  The drive copy is copying data in the disk type storage device 300 to a spare disk type storage device 300 when a failure of one disk type storage device 300 is predicted. In this case, the disk control unit 140A reads the data in the logical chunk 410b on the disk type storage device 300 where a failure is predicted, and in that case, executes the logical storage area release processing in the LDEV pool 68.

  (4) Fourth specific example of this embodiment.

  The fourth specific example is an example where a normal LU and a virtual LU coexist.

  FIG. 17 shows a processing flow of the channel control unit 110C when a write request is received from the host terminal 110C in the fourth specific example.

  The channel control unit 110C determines whether or not the write request received from the host terminal 200C is for the virtual LU 310C (S500).

  When a positive determination result is obtained as a result of the determination in S500 (YES in S500), the channel control unit 110C executes the processes of S510 to S580 described with reference to FIG.

  On the other hand, if a negative determination result is obtained as a result of the determination in S500 (NO in S500), in other words, if it is determined that the write request is for a normal LU, the CHP 119 of the channel control unit 110C A cache area sufficient to store host data is secured on the CM 130 (S590). Then, the CHP 119 stores the host data in the secured cache area (S591).

  FIG. 18 shows a processing flow of the channel control unit 110C when a read request is received from the host terminal 110C in the fourth specific example.

  The channel control unit 110C determines whether or not the read request received from the host terminal 200C is for the virtual LU 310C (S600).

  If a positive determination result is obtained as a result of the determination in S600 (YES in S600), the channel control unit 110C executes the processes in S601 to S610 described with reference to FIG.

  On the other hand, if a negative determination result is obtained as a result of the determination in S600 (NO in S600), in other words, if it is determined that the read request is for a normal LU, the CHP 119 of the channel control unit 110C Then, it is determined whether or not a cache area sufficient to store host data is secured on the CM 130 (S611). If the cache area has been completed (YES in S611), the CHP 119 instructs the disk controller to store the host data in the cache area (S613). On the other hand, if the cache area has not been secured (NO in S611), A cache area is secured (S612), and S613 is executed. Thereafter, the CHP 119 transfers host data from the cache area to the host terminal 200C (S614).

  According to the fourth specific example, processing is executed based on whether the I / O request issuance destination is a normal LU or a virtual LU.

  (5) A fifth specific example of the present embodiment.

  FIG. 19 shows a configuration and a processing flow according to the fifth specific example.

  In the fifth specific example, the control unit 323 (OS 472) of the host terminal 200C transmits a system information request command (S701). This system information request command is, for example, an Inquiry command or a device discovery command according to the SCSI protocol. For example, when the power of the host terminal 200C is turned on or when the OS 472 is restarted, the control unit 323 transmits an Inquiry command. For example, when the power of the host terminal 200C is turned on or when a predetermined instruction is received from the user via the user interface 474, the control unit 323 transmits a device discovery command.

  The channel control unit 110C that has received the system information request command acquires system information from the LU-LDEV management table 162b registered in the SM 120, and transmits the system information to the host terminal 200C. The system information transmitted to the host terminal 200C is registered in the memory 321 of the host terminal 200C. The system information is information associated with each LUN, for example, and includes at least an LDEV attribute associated with the LUN.

  When transmitting an I / O request to the channel control unit 110C, the control unit 323 requests an I / O request to either the normal LU 310B or the virtual LU 310C based on the LDEV attribute included in the system information registered in the memory 321. Is selected (S703).

  Specifically, for example, the control unit 323 displays the attributes of a plurality of LUs that can be accessed by the host terminal 200C based on the LDEV attributes registered in the memory 321, and makes an I / O request to which LU to the user. It may be selected whether to issue, and the LU selected by the user may be selected as an I / O request issue destination LU. In addition, for example, the control unit 323 refers to the host data attribute (for example, metadata of the host data) stored in advance in the memory 321 or the like, grasps the data attribute of the host data to be written or read, and Based on the data attribute and the LDEV attribute, it may be automatically selected whether the I / O request is issued to the normal LU 310B or the virtual LU 310C. More specifically, for example, the control unit 323 selects the normal LU 310B in the case of host data that requires high reliability, and selects the virtual LU 310C in the case of host data that requires high-speed data access. To do.

  The control unit 323 transmits an I / O request to the LU selected in S703 (S704).

  According to the fifth specific example, the host terminal 200C can identify which is a normal LU and which is a virtual LU, and can select a normal LU and a virtual LU.

  (6) Sixth specific example of the present embodiment.

  FIG. 20 shows a configuration and a processing flow related to the sixth specific example.

  The storage controller 100 forms an LU pair composed of two LUs, and performs snapshots for copying data in the primary LU 310P to the secondary LU 310S with one LU as the primary LU 310P and the other LU as the secondary LU 310S. ing.

  Specifically, for example, the storage control device 100 manually sets the normal LU 310A to a request from an external terminal (for example, the SVP 160) or automatically when an I / O request indicating a write request is received (S51). An LU pair is formed with the primary LU 310P and the virtual LU 310C as the secondary LU 310S.

  Thereafter, the storage controller 100 identifies the write destination chunk (for example, the head LBA) 410P in the normal LU 310A from the received write request (S52). Then, the storage control device 100 reads out all the data in the chunk 410P even if only a part of the data in the specified chunk 410P (the portion indicated by the hatched line) is updated (S53).

  At that time, the data check circuit 67 in the storage control device 100 compares whether the data conforms to the non-write target data pattern (S54).

  If a negative comparison result is obtained as a result of the comparison, the storage control device 100 writes the data in the virtual LU 310C (S55). Specifically, as described above, the storage control device 100 dynamically associates the virtual chunk 410S in the virtual LU 310C with the logical chunk in the LDEV pool 68, and associates the data with the virtual chunk 410S. Write to the attached LDEV chunk.

  After creating the snapshot in this way, the storage control device 100 writes the data included in the write request received in S51 into the specified chunk 410P (S56).

  On the other hand, when a negative comparison result is obtained as the comparison result in S54, the storage control device 100 deletes the data read in S53 (S57).

  In the above flow, when the storage control device 100 attempts to form an LU pair, it refers to the LU-LDEV management table 162b and creates an LU pair only when the storage capacity values of the primary LU 310P and the secondary LU 310S match. If they are formed and they do not match, the LU pair formation may be rejected.

  When the LU pair is not formed unless the storage capacity values of the primary LU 310P and the secondary LU 310S are the same, if the storage capacity of the primary LU 310P is large, the storage capacity of the secondary LU 310S must also be large. is there. However, in snapshot creation, the data size to be copied from the primary LU 310P to the secondary LU 310S may be smaller than the capacity of the primary LU 310P. Nevertheless, if the large-capacity normal LU 310A is used as the secondary LU 310S, a large amount of useless free storage areas may occur.

  However, as described above, in creating a snapshot, by forming an LU pair with the secondary LU 310S as the virtual LU 310C, the free space in the LDEV pool 68 that can be associated with the virtual LU 310C is dynamically allocated to another virtual LU 310C. Since they are associated with each other, it is possible to prevent an unnecessary storage area from being generated.

  FIG. 21 shows a specific processing flow related to the sixth specific example.

  For example, in the memory (for example, SM 120) in the storage control subsystem 102, a snapshot management bitmap 794 is registered for each LU (or each LDEV). In the snapshot management bitmap 794, information (for example, 1 or 0) indicating whether the snapshot has been completed is registered for each chunk of the corresponding LU (or LDEV).

  When receiving the write request, the CHP 119 of the channel control unit 110C refers to the snapshot management bitmap 794 corresponding to the primary LU 310P that is the target of the write request, and whether or not the snapshot has been created for the logical chunk to which data is written. Is discriminated (S651).

  If the snapshot has been created as a result of S651 (YES in S651), the CHP 119 stores the host data included in the write request in the logical chunk 410P of the primary LU 310P (S660).

  On the other hand, if the snapshot has not been created as a result of S651 (NO in S651), the CHP 119 secures a cache area for the copy source (eg, logical chunk 410P) and stores the copy source data in the cache area (S653). ).

  Thereafter, the CHP 119 secures a cache area of the copy destination (for example, the virtual chunk 410S) (S654), activates the data check circuit 67, and stores the data in the cache area in the copy destination (S656).

  At that time, the data check circuit 67 compares whether or not the data pattern of the data stored from the copy source to the copy destination matches the non-write target data pattern.

  If a negative comparison result is obtained (NO in S657), the CHP 119 executes S660 described above.

  On the other hand, if a positive comparison result is obtained (YES in S657), the CHP 119 discards the cache area reserved for the copy destination and erases the data stored in the copy destination (S658). . Also, the CHP 119 accesses the snapshot management bitmap 794, and updates the information corresponding to the copy source so that the snapshot has been acquired (S659).

  As mentioned above, although embodiment of this invention was described, these are the illustrations for description of this invention, Comprising: It is not the meaning which limits the scope of the present invention only to this embodiment. The present invention can be implemented in various other forms. For example, in the above description, the data pattern comparison by the data check circuit 67 is performed only on the data stored in the virtual LU 310C (virtual LDEV 61c), but is not limited thereto, and is stored in the normal LU 310A, 310B (normal LDEV 61a). (I.e., if the comparison yields a positive result, for example, the data may normally be deleted without being stored in the LDEV or after being stored) ). Further, for example, a snapshot may be created after a pair of virtual LUs is formed (that is, data copy may be performed between logical chunks in one or a plurality of LDEV pools 68). Further, for example, the virtual LU may be the primary LU and the normal LU may be the secondary LU.

DESCRIPTION OF SYMBOLS 1 ... Storage system, 2 ... RAID group, 10 ... Basic enclosure, 12 ... Expansion enclosure, 100 ... Storage controller, 101 ... Storage device unit, 102 ... Storage control subsystem, 110A-110D ... Channel control part, 120 ... Shared memory, 130 ... Cache memory, 140A to 140D ... Disk controller, 160 ... Maintenance terminal, 200A to 200D ... Host terminal, 300 ... Disk type storage device, 310 ... Logical unit, 600 ... Storage control system

Claims (20)

In the storage control system connected to the host device,
A plurality of channel adapters connected to the host device and receiving data from the host device;
One or a plurality of memories connected to the plurality of channel adapters and storing data exchanged with the host device and control information related to the data;
A plurality of disk adapters connected to the memory and controlling the data to be written to or read from the memory;
A plurality of disk drives connected to the plurality of disk adapters, wherein the data is stored under the control of the plurality of disk adapters;
The first channel adapter included in the plurality of channel adapters is configured to transfer a plurality of first logical volumes generated based on storage areas of the plurality of first disk drives included in the plurality of disk drives to the host device. And when a data write access is made from the higher-level device to the first logical volume selected from the plurality of first logical volumes, an area reserved in advance in the memory is provided. To write the data,
The second channel adapter included in the plurality of channel adapters collectively manages a plurality of second logical volumes generated based on storage areas of the plurality of second disk drives included in the plurality of disk drives, When the information about the third logical volume provided to the host device is received from the host device, if there is a write access of the data from the host device to the third logical volume, the third logical volume The area on the memory required as the second logical volume corresponding to the volume is secured for the area corresponding to the data, and the data is written to the reserved area, corresponding to the third logical volume. The second logical volume is released,
Memory control system. At least one of the one or more memories stores one or more non-write target data patterns that are data patterns of non-write target data that should not exist in the second logical volume;
At least one of the second channel adapter and the plurality of disk adapters may be configured such that a data pattern of the data is at least one of the one or more non-write target data patterns stored in the memory. Perform a data pattern comparison of whether or not it matches, and if the result of the comparison is positive, discard the data;
The storage control system according to claim 1. The non-write target data pattern is a format data pattern when the data is received from the higher-level device in block units that are data storage units of the second logical volume.
The storage control system according to claim 2. At least one of the second channel adapter and the plurality of disk adapters includes a processor and a data check unit having a register,
The data check circuit unit performs a data pattern comparison as to whether or not a data pattern of the data matches at least one of the one or more non-write target data patterns stored in the memory; Write the result of the comparison to the register,
The processor obtains the result of the data pattern comparison from the register and discards the data if the result of the data pattern comparison is positive;
The control system according to claim 2. A maintenance terminal for forming a plurality of logical paths connecting the first logical volume and the third logical volume and the higher-level device;
The storage control system according to claim 1. At least one of the second channel adapter and the plurality of disk adapters, when there is a write access, a third logical storage area in the third logical volume and a second logical volume in the second logical volume Performing association with a storage area, and releasing the association by releasing the association if the result of the data pattern comparison is positive,
The storage control system according to claim 1. When at least one of the second channel adapter and the plurality of disk adapters acquires the data read from the third logical volume via the second logical volume, the data pattern of the data is , Performing a data pattern comparison as to whether or not it matches at least one of the one or more non-writeable data patterns stored in the memory, and if the result of the comparison is affirmative, Erase the data present in the two logical volumes;
The storage control system according to claim 1. The second logical volume is composed of a plurality of logical chunks,
At least one of the plurality of disk adapters has a data size of the data stored in the logical chunk even if the data pattern comparison result for the data when the write access is made is positive. If the data is smaller than the capacity, the data is stored in the second logical volume, and when data is read, the data is read in units of the logical chunk, and a positive data comparison result is obtained for the read data. If read, the read data is deleted from the logical chunk.
The storage control system according to claim 7. When the data is written to the first logical storage area of the first logical volume, at least one of the second channel adapter and the plurality of disk adapters reads the data in the first logical storage area; The data pattern of the read data is copied to the second logical storage area of the second logical volume, and the read data pattern is one of the one or more non-write target data patterns stored in the memory. If the result of the comparison is affirmative, discard the read data so that it does not exist in the second logical storage area.
The storage control system according to claim 1. At least one of the one or more memories stores one or more non-write target data patterns that are data patterns of non-write target data that should not exist in the second logical volume;
The second channel adapter is adapted to associate the third logical storage area in the third logical volume with the second logical storage area in the second logical volume, and read access from the higher-level device When the access is to a third logical storage area that is not associated with the second logical storage area, at least one of the one or more non-write data patterns is transmitted to the higher-level device.
The storage control system according to claim 1. At least one of the one or more memories stores attributes of the first and third logical volumes;
The second channel adapter reads the attributes of the first and third logical volumes from the memory and notifies the host device;
The storage control system according to claim 1. In the control method of the storage control system connected to the host device,
The storage control system comprises:
A plurality of channel adapters connected to the host device and receiving data from the host device;
One or a plurality of memories connected to the plurality of channel adapters and storing data exchanged with the host device and control information related to the data;
A plurality of disk adapters connected to the memory and controlling the data to be written to or read from the memory;
A plurality of disk drives connected to the plurality of disk adapters, wherein the data is stored under control of the disk adapters;
Providing a plurality of first logical volumes generated based on storage areas of a plurality of first disk drives included in the plurality of disk drives to the host device;
A step of writing the data to an area reserved in advance in the memory when there is a write access of the data from the host device to the first logical volume selected from the plurality of first logical volumes; ,
Collectively managing a plurality of second logical volumes generated based on storage areas of a plurality of second disk drives included in the plurality of disk drives;
When the information about the third logical volume provided to the higher-level device is received from the higher-level device, if there is a write access of the data from the higher-level device to the third logical volume, the third logical volume The area on the memory required as the second logical volume corresponding to the logical volume is secured for the area corresponding to the data, the data is written to the secured area, and the third logical volume is written. Releasing the corresponding second logical volume. Storing at least one of the one or more memories one or more non-write data patterns that are data patterns of non-write data that should not exist in the second logical volume;
A data pattern comparison is performed to determine whether the data pattern of the data matches at least one of the one or more non-write target data patterns stored in the memory, and the result of the comparison is positive 13. The method of claim 12, further comprising: discarding the data. The non-write target data pattern is a format data pattern when the data is received from the higher-level device in block units that are data storage units of the second logical volume.
The method of claim 13.   13. The method according to claim 12, further comprising the step of forming a plurality of logical paths connecting the first logical volume and the third logical volume and the host device. Correlating the third logical storage area in the third logical volume with the second logical storage area in the second logical volume when there is a write access;
The method according to claim 12, further comprising: releasing the association and releasing the association when a result of the data pattern comparison is positive. When the data read from the third logical volume is acquired via the second logical volume, the data pattern of the data is one of the one or more non-write target data patterns stored in the memory. Performing a data pattern comparison of whether or not it fits at least one of them,
The method of claim 12, further comprising: erasing the data present in the second logical volume if the result of the comparison is positive. When the data is written to the first logical storage area of the first logical volume, the data in the first logical storage area is read and copied to the second logical storage area of the second logical volume;
Performing a data pattern comparison as to whether or not a data pattern of the read data matches at least one of the one or more non-write data patterns stored in the memory;
The method according to claim 12, further comprising: discarding the read data so that it does not exist in the second logical storage area if the result of the comparison is positive. Storing at least one of the one or more memories one or more non-write data patterns that are data patterns of non-write data that should not exist in the second logical volume;
Associating a third logical storage area in the third logical volume with a second logical storage area in the second logical volume;
When the read access from the host device is an access to a third logical storage area that is not associated with the second logical storage area, at least one of the one or more non-write target data patterns is set to the The method according to claim 12, further comprising: transmitting to a host device. Storing the attributes of the first and third logical volumes in at least one of the one or more memories;
The method according to claim 12, further comprising: reading out the attributes of the first and third logical volumes from the memory and notifying the higher-level device.