WO2016103416A1 - Storage system, storage device and access control method - Google Patents

Abstract

A storage system that connects one or more host calculators and a storage device, wherein said storage device comprises first layer logical units (LUs) and second layer LUs that are mapped hierarchically to each other, a configuration control unit for dividing the second layer LUs into one or more logical groups and mapping the logical groups to the first layer LUs, and an input/output control unit for receiving an access request in which the first layer LUs and the second layer LUs are specified and limiting the access to be processed among the accesses to the specified second layer LUs to the access to the second layer LUs divided into the logical groups that are mapped to the specified first layer LUs, and said host calculator transmits the access request to the storage device.

Description

Storage system, storage device, and access control method

The present invention relates to a storage system, a storage apparatus, and an access control method.

With the progress of virtualization technology for host computers, a plurality of virtual servers are constructed on a single physical host computer, and a configuration is used in which images of virtual servers and usage data are stored in a storage device. With this configuration, the data of one or a plurality of virtual servers is collected in one logical volume of the storage device.

Patent Document 1 discloses a technology in which one logical volume of a storage device is divided into a plurality of virtual volumes on a host computer, and data used by a virtual server is stored for each virtual volume.

Patent Document 2 discloses a technology that accepts only a request from a WWN (World Wide Name) registered at the time of fabric login as LU access restriction via In-band such as SAN (Storage Area Network). Yes.

JP 2012-079245 A JP 2004-227154 A

In a storage device, a plurality of logical volumes (LU: Logical Unit) are grouped into a plurality of logical groups (LU Configurate), and a logical path is set for a representative logical volume (ALU: Administrative Logical Unit) in the logical group. And The host computer issues an I / O (Input / Output) command specifying an identifier of a logical volume (SLU: Subsidy Logical Unit) other than the ALU in the logical group to which the logical path belongs, and the storage device receives the received I / O. Distributes I / O processing to the SLU specified by the command. Hereinafter, such a configuration of the storage apparatus is referred to as “Conglomerate LU Structure”.

In a storage device of Conglomerate LU Structure, it is possible to assign a logical volume of the storage device in units of virtual servers by assigning one or more SLUs to one virtual server on the host computer.

The plurality of virtual volumes of the technique disclosed in Patent Document 1 is one logical volume in the storage apparatus, and the functions of the storage apparatus are controlled in units of logical volumes. Therefore, in the configuration of the storage device disclosed in Patent Document 1, for example, when the local copy function or the remote copy function provided in the storage device is used, the logical volume divided into a plurality of virtual volumes becomes a copy unit. An extra cost is incurred for copying the data of the virtual volume that does not need to be copied. In Patent Literature 1 and Patent Literature 2, the Conglomerate LU Structure is not assumed, so the access to the logical volume in which the logical path is set in the storage device is limited, and the SLU that is not the ALU in which the logical path is set. The technology for restricting access is not disclosed.

Therefore, an object of the present invention is to make it possible to restrict access to SLUs in a hierarchical configuration of ALUs and SLUs.

In order to solve the above problems, a typical storage system according to the present invention is a storage system in which one or a plurality of host computers and storage devices are connected, and the storage devices are associated in a hierarchical manner. A configuration controller that includes a first layer LU and a second layer LU, distributes the second layer LU to one or more logical groups, and associates the logical group with the first layer LU; Receiving the access request in which the first layer LU and the second layer LU are designated from the host computer, and distributing the access request to the logical group associated with the designated first layer LU. The access to the second layer LU is provided with an input / output control unit for limiting access to be processed in the access to the designated second layer LU.

Further, in the storage system according to the present invention, the storage device is connected to the host computer, and host computers that can access the LU of the first layer are registered, and only access from the registered host computer is performed. A connection I / F to be permitted is further provided.

According to the present invention, it is possible to restrict access to SLUs in a hierarchical configuration of ALUs and SLUs.

It is a figure which shows the example of a structure of a storage system. It is a figure which shows the example of a structure of a storage apparatus. It is a figure which shows the example of a structure of a host computer. It is a figure which shows the example of a structure of a management computer. It is a figure which shows the example of a structure of a logical port management table. It is a figure which shows the example of a structure of a logical volume management table. It is a figure which shows the example of a structure of a binding management table. It is a figure which shows the example of a structure of a storage apparatus management table. It is a figure which shows the example of a structure of a host computer management table. It is a figure which shows the example of a Report LUN command. It is a figure which shows the example of the response of Report LUN command. It is a figure which shows the example of LUN specification of a SCSI command. It is a figure which shows the example of the flowchart which processes a SCSI command. It is a figure which shows the example of the flowchart of a process which produces ALU. It is a figure which shows the example of the flowchart of the process which allocates ALU to a host group. It is a flowchart of the process which determines the allocation to a new host group. It is a figure which shows the example of the path | route from a logical port to SLU. It is a figure which shows the example of the flowchart of the process which extracts allocated ALU. It is a figure which shows the example of the flowchart of a process which determines the new path | route 1. FIG. It is a figure which shows the example of the flowchart of the process which determines the new paths 2 and 3. FIG. It is a figure which shows the example of the flowchart of a process which binds SLU to the designated ALU. It is a figure which shows the example of the flowchart of the process which extracts ALU allocated to the request origin host computer. It is a figure which shows the example of the flowchart of a process which detects ALU to which the bind group was allocated from the extracted ALU. It is a figure which shows the example of the flowchart of a process which binds SLU to the detected ALU. It is a figure which shows the example of a structure of the storage system which has a host computer which operates bind. It is a figure which shows the example of a structure of the host computer which operates bind.

Hereinafter, examples will be described with reference to the drawings.

In the following description, the information of each embodiment will be described using the expression “table”. However, the information is not necessarily a table, and may be expressed by a data structure other than a table. In some cases, “program” is the subject of the description, but the program is executed by the processor, and the processing determined using the memory and communication port is the same as that for the processor as the subject. .

Further, the processing obtained by executing the program by the processor is the same as the processing by the dedicated hardware, and part or all of the processing may be realized by the dedicated hardware. For this reason, an object is described using the expression “unit”. However, a part or all of these “units” may be a program executed by a processor, or a part or all of them are dedicated hardware. There may be.

Also, the program may be installed by a program distribution server or a computer-readable storage medium.

FIG. 1 is a diagram illustrating an example of the configuration of a storage system according to the first embodiment. The storage apparatus 2000 includes a logical partitioning unit that divides a logical volume into a plurality of logical partitions (hereinafter, BG: bind group 2600), an I / O control unit (input / output control unit) corresponding to the Congregate LU Structure, A logical volume 2210 is provided from the logical port 2110 to the host computer 1000 connected via the SAN 4000.

Here, the logical port 2110 is a generic name in the case where the logical ports (I / F: InterFace) 2110a, 2110b, etc. are represented without distinction, and in the following description, the alphabetical symbols are omitted to represent the generic name. Further, the logical volume 2210 is a generic name for the ALUs 2210a, 2210b, and the SLUs 2210c to 2210g, which are represented without distinction. The ALU 2210h is a generic name for representing the ALUs 2210a, 2210b, etc. without distinction. The SLU 2210i is a generic name for representing each of the SLUs 2210c to 2210g without distinction.

The storage apparatus 2000 is connected to the management computer 3000 via a management network 5000 represented by a LAN (Local Area Network). The storage apparatus 2000 provides the management computer 3000 with each configuration management table provided in the storage apparatus 2000 and a management API (Application Program Interface) provided by the configuration control unit.

In FIG. 1, an ALU 2210a and an ALU 2210b whose ALU attributes are set by the storage administrator exist as two logical volumes 2210. One BG (bind group) 2600a is assigned to the ALU 2210a by the storage administrator, and one ALU 2210b is assigned to the ALU 2210b. Two BG2600b are allocated. Three SLUs 2210c to 2210e are assigned to the BG 2600a, and two SLUs 2210f and 2210g are assigned to the BG 2600b by the storage administrator. Further, the storage administrator assigns ALU 2210a to HGs (host groups) 2111a and 2111c included in the logical port 2110a, and assigns ALU 2210b to HGs 2111b and 2111d.

Further, the HG 2111a and 2111c register a WWN for identifying an HBA (Host が Bus Adapter) included in one or a plurality of host computers 1000a belonging to the business A. The HGs 2111b and 2111d are registered with the business B. The WWN of the HBA that one or more host computers 1000b to which it belongs is registered. Then, for example, three virtual servers operate on the host computer 1000a belonging to the job A, and SLU 2210c, SLU 2210d, and SLU 2210e are assigned to each of the three virtual servers. Further, for example, two virtual servers operate on the host computer 1000b belonging to the business B, and each of the two virtual servers can be assigned SLU 2210f and SLU 2210g, respectively. Each SLU 2210 stores data (virtual server image data, usage data, etc.) of the virtual server to which each SLU 2210 is assigned.

The management computer 3000 is connected to the host computer 1000 via the management network 5000, can acquire configuration information of the host computer 1000, and can receive a storage management operation request from the host computer 1000. A storage administrator who manages the storage apparatus 2000 can recognize the configuration information of the storage apparatus 2000 and the host computer 1000 via the storage apparatus management program 3110, and further transmits a configuration change instruction to each apparatus. be able to.

Hereinafter, setting of ALU, SLU, and BG by the storage apparatus management program 3110 and I / O control in the storage apparatus 2000 based on the setting will be described. First, the configuration of each device will be described.

FIG. 2 is a diagram showing an example of the configuration of the storage apparatus 2000. The storage apparatus 2000 includes an FEPK (Front End PacKage) 2100 that is a host I / F unit, an MPPK (Micro Processor PacKage) 2200 that is a control unit, a CMPK (Cache Memory PacKage) 2300 that is a shared memory unit, and a disk I / F. One or a plurality of BEPKs (BackEnd PacKage) 2400, an HDD (Hard Disk Drive) 2700 as an example of a storage device, and a management server 2500 are provided.

The internal network 2800 connects the FEPK 2100, MPPK 2200, CMPK 2300, BEPK 2400, and the management server 2500 to each other. Through the internal network 2800, each MP 2240 of the MPPK 2200 can communicate with any of the FEPK 2100, CMPK 2300, and BEPK 2400.

The FEPK 2100 has a plurality of logical ports 2110 each serving as a host I / F. Further, the logical port 2110 has a plurality of host groups 2111, and only a login request from the WWN registered in the host group 2111 at the time of login processing to the logical port 2110 via the SAN 4000 (usually referred to as fabric login). It has a function to accept.

The BEPK 2400 has a plurality of disk I / Fs 2410. The disk I / F 2410 is connected to the HDD 2700 via, for example, a cable, and is also connected to the internal network 2800, and mediates a data transfer process between the internal network 2800 side and the HDD 2700 to be read or written. .

The CMPK 2300 includes a data cache memory 2310 and a control information memory 2320. The cache memory 2310 and the control information memory 2320 may be volatile memories, for example, DRAM (Dynamic Random Access Memory). The cache memory 2310 temporarily stores (caches) data to be written to the HDD 2700, or temporarily stores (caches) data read from the HDD 2700. The control information memory 2320 stores a logical port management table T1000, a logical volume management table T2000, and a bind management table T3000 as information necessary for control, for example, configuration information of the logical volume 2210 and BG2600.

The MPPK 2200 includes a plurality of MPs (Micro Processors) 2240, an LM (Local Memory) 2220, and a bus 2230 that connects these. The MP 2240 is a processor that is also used in a computer or the like, and becomes a logical partition unit, an I / O control unit, a configuration control unit, or the like by operating according to a program stored in the LM 2220. The LM 2220 stores a part of control information for I / O control stored in the control information memory 2320.
The management server 2500 is a computer having an application for transmitting an operation request from the management computer 3000 to a control program loaded from the control information memory 2320 to the LM 2220 and executed by the MP 2240. The management server 2500 can also have each program that the management computer 3000 has.

The control information memory 2320 stores information handled by the logical partitioning unit and the I / O control unit. The storage device management program 3110 on the management computer 3000 can acquire information stored in the control information memory 2320 via the management server 2500.

The logical partitioning unit distributes a logical volume 2210 (also referred to as LDEV: Logical Device), which is a logical storage area provided by the BEPK 2400, to the BG 2600. The logical partitioning unit assigns an identifier (BIND GROUP ID) that can uniquely identify the BG2600 in the device to the configuration information of the logical volume 2210 registered in the BG2600, and stores it in the control information memory 2320 as the logical volume management table T2000. To do.

FIG. 3 is a diagram showing an example of the configuration of the host computer 1000. The host computer 1000 is a computer including a processor 1100, a memory 1200, an HBA (HostＨBus） Adapter) 1300, a Hyper Visor unit 1400, an output unit 1600, an input unit 1700, a management port 1800 that is a network I / F, and the like. The host computer 1000 is, for example, a personal computer, a workstation or a main frame. The memory 1200 stores a business application 1210, a storage management program 1220, and an OS (Operating System).

The Hyper-Visor unit 1400 is a resource (logical processor, logical memory, logical network I / F, etc.) obtained by abstracting hardware resources (hardware resources: processor 1100, memory 1200, network I / F 1800, etc.) of the host computer 1000. And the abstracted resource is allocated to a virtual server (virtual machine). Alternatively, the Hyper-Visor unit 1400 logically divides the hardware resources (processor 1100, memory 1200, network I / F 1800, etc.) included in the host computer 1000, and divides the logically divided hardware resources into virtual servers. (Logical partition: Logical Partition) may also be assigned.

Each virtual server operates on the host computer 1000 based on the resources allocated by the Hyper-Visor unit 1400, and executes the OS and the business application 1210.

The processor 1100 performs overall control of the host computer 1000, and operates according to each program by executing the business application 1210 stored in the memory 1200 or a program of an alternate path (not shown). For example, the processor 1100 issues a read access request or a write access request as an access request to the storage apparatus 2000 by executing the business application 1210. The memory 1200 is used for storing programs and the like, and is also used as a work memory for the processor 1100.

The HBA 1300 performs protocol control during communication with the storage apparatus 2000. When the HBA 1300 performs protocol control, data and commands according to, for example, a fiber channel protocol are transmitted and received between the host computer 1000 and the storage apparatus 2000.

The raw device 1310 corresponds to the ALU 2210h of the storage apparatus 2000, and is a device that substitutes for access to the ALU 2210h in the host computer 1000. The virtual volume 1320 corresponds to the SLU 2210 i of the storage apparatus 2000 and is a device that substitutes for access to the SLU 2210 i in the host computer 1000. The raw device 1310 is a device for reducing overhead by executing I / O processing without temporarily copying data to a page cache (not shown) when the Hyper-Visor unit 1400 is accessed. Used as

The volume identifier acquired by the HBA 1300 needs to be a globally unique value, and is formed by a combination of the serial number (device serial number) of the storage device 2000 and the local volume identifier of the logical volume 2210 in the storage device 2000, for example. In the first embodiment, the storage system 2000 has a function of responding to the identifier of the ALU 2210h in response to the response of the volume identifier.

The Hyper-Visor unit 1400 uses the SLU 2210i associated with the ALU 2210h corresponding to the raw device 1310 as one virtual volume 1320, and each application such as an OS (Operating System) or a business application 1210 executed on the host computer 1000 or the virtual server. To provide.

The Hyper-Visor unit 1400 extracts the corresponding ALU 2210h and SLU 2210i for the I / O command to the virtual volume 1320 issued from each application or OS, and assigns the identifiers of the ALU 2210h and SLU 2210i to the I / O command. A command is issued to the raw device 1310.

The input unit 1700 includes, for example, a keyboard, a switch, a pointing device, a microphone, and the like. The output unit 1600 includes a monitor display and a speaker.

In addition, the storage management program 1220 transmits an operation request for the logical volume 2210 such as SLU 2210i allocation or deallocation from the business application 1210 or the OS to the storage apparatus management program 3110 included in the management computer 3000 via the management network 5000. be able to. Furthermore, the storage management program 1220 can also send a logical volume operation request to the configuration control unit of the storage apparatus 2000 via the SAN 4000.

FIG. 4 is a diagram showing an example of the configuration of the management computer 3000. The management computer 3000 includes a memory 3100, a processor 3200, a storage medium 3300, a network I / F 3400, an output unit 3600, and an input unit 3700. In the management computer 3000, the processor 3200 performs overall control of the management computer 3000 and executes the storage device management program 3110 or the host computer management program 3120 loaded in the memory 3100, whereby various storage devices and host computer configurations are configured. The management operation request is transmitted to the storage apparatus 2000 and the host computer 1000 via the network I / F 3400 and the management network 5000.

The memory 3100 stores a storage device management table T4000 and a host computer management table T5000 as control information used by the storage device management program 3110 and the host computer management program 3120. The storage administrator can input an operation request to the input unit 3700 such as a keyboard or a mouse, and can acquire an execution result via the output unit 3600 such as a display or a speaker. A storage medium 3300 such as an HDD or SD stores a storage device management program 3110, a host computer management program 3120, and an execution log of each control program.

FIG. 5 is a diagram showing an example of the logical port management table T1000 provided in the storage apparatus 2000. The logical port management table T1000 is referred to during I / O control processing, ALU allocation processing, and bind group allocation processing executed by the storage apparatus 2000, and further managed via the management network 5000 by the API provided by the management server 2500. It is also accessed from the storage device management program 3110 provided in the computer 3000.

The logical port management table T1000 includes a column T1010 for registering an identifier that can identify the logical port 2110 in the apparatus, a column T1020 for registering the WWN of the logical port 2110, and a host group 2111 included in the logical port 2110. Registers a field T1030 for registering an identifier uniquely identifiable within a local range under 2110, and a WWN (HOSTＨWWN) of a connection source (host computer or a virtual server of the host computer) that permits fabric login in the host group 2111. A column T1040 and a column T1050 for registering the unique identifier of the ALU assigned to the host group 2111 in the apparatus.

By referring to the logical port management table T1000, the storage system 2000 enables fabric login to the logical port 2110 from the connection source (host computer or virtual server) having HOST WWN registered in the column T1040. The connection source (host computer or virtual server) having HOST WWN registered in T1040 can recognize the ALU registered in the column T1050. Here, the LUN (Logical Unit Number) in the column T1050 is the ID of the raw device 1310, and the ALU ID is an identifier in the storage apparatus 2000, whereas the LUN is an identifier used by the host computer 1000. Therefore, a plurality of different host computers can use the same LUN (for example, 243).

FIG. 6 is a diagram showing an example of the logical volume management table T2000 included in the storage apparatus 2000. The logical volume management table T2000 is referred to and registered by a logical volume configuration control unit and a bind group configuration control unit (both not shown) included in the storage apparatus 2000. These control units are provided by the management server 2500. The information of the logical volume management table T2000 is provided based on the request from the API to be updated, and the registration information is updated.

The logical volume management table T2000 includes a column T2010 for registering an identifier that can uniquely identify a logical volume (LDEV) within the apparatus, a column T2020 for registering the capacity of the logical storage area of the LDEV, and a physical storage area to which the LDEV belongs. A field T2030 for registering an identifier uniquely indicated in the apparatus, a T2040 for registering an LDEV attribute, and a T2050 for registering an identifier uniquely indicating the bind group to which the LDEV belongs in the apparatus.

The column T2040 may include information indicating an ALU or information indicating an SLU. Since the ALU is not an LDEV that actually stores data, there is no physical storage area information in the ALU column T2030, and the capacity of the ALU column T2020 is 0. The column T2030 may include an identifier of a parity group of RAID (Redundant Arrays of Inexpensive Disks) as a physical storage area.

FIG. 7 is a diagram illustrating an example of the bind management table T3000 included in the storage apparatus 2000. The bind management table T3000 is registered by a bind configuration control unit (not shown) included in the storage apparatus 2000, and is referred to during I / O control. The bind management table T3000 includes a column T3010 for registering an identifier that can uniquely identify an ALU within the apparatus (LDEV ID is used in the example of FIG. 7), and a column for registering a bitmap indicating a list of LDEVs that are in a bound state. It consists of T3020.

In the column T3020, the values 0 to 16777215 are LDEV IDs, and each bit T3025 in the column below corresponds to one SLU. For example, if LDEV ID = 00005 is bound to ALU ID = 000000 as an SLU, the sixth bit from the right is 1 in the bind management table T3000 shown in FIG. This bit T3025 is registered by the bind configuration control unit of the storage apparatus 2000 and is referred to during I / O control.

FIG. 8 is a diagram showing an example of the storage device management table T4000 included in the management computer 3000. The storage device management table T4000 is registered when the storage device management program 3110 included in the management computer 3000 detects the storage device 2000 connected to the management network 5000. This detection timing is a manual detection timing or an automatic detection timing by SLP (ServiceＳLocation Protocol).

The storage device management table T4000 includes a column T4010 for registering an identifier that can uniquely identify the storage device 2000 under the storage device management program 3110, a column T4020 for registering an identifier that can uniquely identify the storage device 2000 globally, The column T4030 is used to register the IP address of the management server 2500 for connecting to the management server 2500 of the storage apparatus 2000 via the management network 5000. The column T4020 may be information configured by a combination of a vendor name, a model, and a manufacturing number.

FIG. 9 is a diagram showing an example of the host computer management table T5000 included in the management computer 3000. The host computer management table T5000 is registered when the host computer management program 3120 detects the host computer 1000 on the management network 5000. This detection is manual detection or automatic detection by SLP. When detecting the host computer 1000, the host computer management program 3120 registers the host name in the column T5010, and registers the IP address of the management port 1800 of the host computer 1000 in the column T5030.

Further, the host computer management program 3120 acquires the configuration information of the host computer 1000 via the management network 5000, and registers the WWN of the HBA 1300 included in the host computer 1000 in the column T5020. Thereafter, the host computer management program 3120, through the storage device management program 2110, based on the HOST HBA WWN registered in the column T5020, the configuration information of the ALU assigned to the HOST HBA WWN from the managed storage device 2000 Are registered in the column T5070 from the column T5040.

Here, the columns T5040 to T5070 are updated periodically or manually by the storage administrator, or when a configuration change notification of the host computer is received from the SLP. In the column T5050, an identifier obtained by combining the identifier of the logical port 2110 registered in the column T1010 of the logical port management table T1000 and the local identifier of the host group 2111 registered in the column T1030 may be registered. .

Next, the I / O control unit included in the storage apparatus 2000 in the Cluster LUN Structure will be described with reference to FIG. 10, FIG. 11, and FIG. First, the REPORT LUN command will be described with reference to FIGS. Conventionally, a SCSI command for acquiring a list of LUNs provided in a logical path from the host computer 1000 via the SAN 4000 has been defined, and this is called a REPORT LUN command.

As shown in FIG. 10, in the “Conglomerate LUN Structure”, the REPORT LUN command C1000 includes a SELECT REPORT code, and the LUN to be acquired can be specified by type (normal LUN, ALU, SLU). For example, when the REPORT LUN command C1000 in which the SELECT REPORT code is designated as ALU is issued from the host computer 1000 to the logical path LUN0, the I / O control unit included in the storage apparatus 2000 accepts this, Referring to port management table T1000, HOST GROUP to which logical path LUN0 belongs is extracted from column T1030, ALU registered in HOST GROUP is extracted from column T1050, and an ALU list is returned in response C2000 shown in FIG. .

Specifically, the number of extracted ALUs is the value of LUN LIST LENGTH of the response C2000, and the LUN given to the extracted ALU (value in parentheses in the column T1050) is LUNＬＵLIST, and the order is returned. The

The relationship between the ALU ID and the LUN of the ALU will be described with reference to FIG. The LUN designation of the data read / write (READ / WRITE) SCSI command is expressed by 2 bytes in the ADMINISTRATION ELEMENT part shown in FIG. In the first embodiment, the ALU ID is expressed by 2 bytes. Since the LUN only needs to be unique within the host group, a single 2-byte ID is assigned to one ALU within the host group. Therefore, a 2-byte ID (LUN) is returned as C2000 LUN LIST.

Next, for example, when the SELECT REPORT code is designated as SLU in the REPORT コ マ ン ド LUN command C1000 toward LUN 0 of the logical path, the I / O control unit performs the same processing as the ALU as already described. Further, referring to the bind management table T3000, the number of SLUs bound to the extracted ALU, that is, the number of bits in which the value of the SLU bitmap T3020 is 1, is set as the value of LUN LIST LENGTH, The LDEV ID corresponding to the bit for which the value of the SLU bitmap T3020 is 1 is sequentially responded as LUN LIST. Here, the LDEV ID is the SLU ID, and the SLU ID corresponds to the SUBSIDIARY ELEMENT part shown in FIG. 12 and is 6 bytes, so the SLU ID is responded as it is.

FIG. 13 is a diagram showing a flowchart F1000 that is an example of READ / WRITE processing to the SLU in the I / O control unit included in the storage apparatus 2000. As described with reference to FIG. 12, the host computer 1000 (or virtual server) transmits a SCSI command specifying the ALU and SLU to the storage apparatus 2000. The logical port 2110 of the storage apparatus 2000 receives the SCSI command, and extracts the designated ALU (designated ALU) and the designated SLU (designated SLU) (step F1010). In the storage apparatus 2000, an I / O control program is loaded into the LM 2220 and transmitted to the MP 2240. The MP 2240 operating as the I / O control unit checks the presence / absence of the designated ALU with reference to the column T1050 of the logical port management table T1000 (step F1020).

If there is no ALU in step F1020, the MP 2240 notifies the requesting host computer 1000 (or virtual server) that there is no ALU (step F1040) and ends the process. When the ALU exists, the MP 2240 refers to the logical volume management table T2000 and extracts the LDEV ID of the designated SLU. The MP 2240 loads the bind management table T3000 into the LM 2220 (Step F1030) and refers to it to determine whether or not the designated SLU is bound to the designated ALU (Step F1050). If not bound, SLU unassigned is notified to the requesting host computer 1000 (or virtual server) (step F1060), and the process is terminated.

If the SLU is bound, the MP 2240 executes designated READ / WRITE processing such as data writing to the logical volume, and returns the result to the requesting host computer 1000 (or virtual server) (step F1070). End the process. As described above, when the designated SLU is bound to the designated ALU, that is, when the bit of the designated SLU for the designated ALU is 1 in the bind management table T3000, the READ / WRITE specified by accepting the command is specified. It is characterized by processing.

The storage apparatus 2000 has a first-layer ALU and a second-layer SLU that are hierarchically associated with each other. Then, the configuration control unit of the storage apparatus 2000 distributes the SLU to one or a plurality of logical groups (bind groups), and associates the bind groups with the ALUs. When the input / output control unit of the storage apparatus 2000 receives an access request in which the ALU and the SLU are designated from the host computer 1000, the input / output control unit processes the access to be processed in the access to the designated SLU. The control is limited to the access to the SLU distributed to the bind group associated with. Therefore, the storage device 2000 can restrict access to the SLU in a hierarchical configuration of the ALU and the SLU (Conglomerate LU Structure).

FIG. 14 is a diagram showing a flowchart F2000 as an example of ALU creation processing in the configuration control unit (not shown) provided in the storage apparatus 2000. In the ALU creation process, the storage device management program 3110 on the management computer 3000 receives an ALU creation instruction from the storage administrator, and sends an instruction to the API provided by the management server 2500. The management server 2500 analyzes the API request, transmits an instruction to the configuration control unit, and the ALU creation process is started after the configuration control unit accepts the instruction.

The MP 2240 operating as the configuration control unit first refers to the logical volume management table T2000 to check whether there is a free LDEVＤＥID (step F2010), and if the LDEV ID is saturated, notifies the management computer 3000 of LDEV shortage. (Step F2020) and the process ends. If there is an empty LDEV ID, the bind management table T3000 is referred to, and it is confirmed that the number of ALUs is less than the upper limit (step F2030). If the number of ALUs is the upper limit, the saturation of the ALU number is sent to the management computer 3000. Notification is made (step F2050) and the process is terminated.

Here, the upper limit value of the number of ALUs may be specified at the time of shipment based on the specifications of the storage device 2000, or may be set in the storage device 2000 by the storage administrator using the storage device management program 3110. Is recorded in the control information memory 2320 in the storage apparatus 2000. If the number of ALUs is less than the upper limit, the MP 2240 selects an empty LDEVＤＥID (step F2040), registers the selected LDEV ID in the column T2010 of the logical volume management table T2000, and stores the selected LDEV ID in the column T2040 indicating the attribute of the LDEV. The attribute is registered (step F2060), and the process is terminated.

Since the data stored in the host computer 1000 is written in the logical storage area indicated by the SLU, the ALU does not have a logical storage area. Therefore, when the column T2040 of the logical volume management table T2000 has an ALU attribute, the value of T2020 indicating the logical storage capacity is 0, and the value of the column T2030 indicating the physical storage area is blank.

Next, ALU allocation processing for making an ALU recognizable from the host computer 1000 (or virtual server) will be described with reference to FIG. FIG. 15 is a diagram illustrating a flowchart F3000 that is an example of ALU allocation processing in the configuration control unit included in the storage apparatus 2000. In the ALU allocation process, the storage administrator inputs an ALU allocation instruction specifying the ALU and host group to the storage apparatus management program 3110 on the management computer 3000, and the storage apparatus management program 3110 instructs the API provided by the management server 2500. The ALU allocation process is started when an ALU allocation instruction arrives from the management server 2500 to the configuration control unit of the storage apparatus 2000.

The MP 2240 operating as the configuration control unit extracts the designated ALU (designated ALU) and the designated host group (designated host group) from the input data (step F3010). Here, the ALU is specified by the ALU ID, and the host group is specified by a pair of PORT ID and HOST ID, that is, the format of the column T5050 shown in FIG. Next, the MP 2240 refers to the logical port management table T1000, extracts the number of existing ALUs for the host group from the column T1050, and confirms that the number of ALUs assigned to the host group is less than the upper limit number. (Step 3030).

Here, the upper limit number of ALUs that can be allocated to the host group is determined by the specifications of the storage apparatus 2000 and is set at the time of shipment of the apparatus, or may be set by the storage administrator using the storage apparatus management program 3110. These are recorded in the control information memory 2320. When the ALU allocation number is the upper limit value, the management computer 3000 is notified that the allocation upper limit number to the host group is exceeded (step F3050), and the process is terminated. If the ALU allocation number is less than the upper limit, PROCESS A (step F3040) shown in FIG. 16 is executed.

PROCESS A is a process for examining the relationship between a host group to which a designated ALU has already been assigned (existing host group) and the designated host group. The host WWN registered in the existing host group is registered in the designated host group. If it is different from the host WWN, it is determined that the designated ALU is assigned to a different host transaction, and the ALU assignment instruction is rejected. Details will be described with reference to FIG.

First, the MP 2240 initializes a variable i serving as a counter to 0 (step F3041), extracts all host groups to which the designated ALU is assigned from the logical port management table T1000, and applies to all the extracted host groups. , The processing from step F3042 to step F3047 is executed.

The MP 2240 extracts the host WWN registered in the extracted host group from the column T1040 of the logical port management table T1000 (step F3042) and compares it with the host WWN registered in the designated host group (step F3043). If the host WWN is different, the counter i is incremented by one (step F3046). If the registered host WWN is the same, the process proceeds to step F3047. When the processing from step F3042 to step F3047 is completed for all the extracted host groups, the value of the counter i is confirmed (step F3048), and if the counter i is greater than 0, it is assigned to a different host transaction. Is sent to the management computer 3000 to prompt the host group to be reviewed (step F3049), and the process is terminated.

If the counter i is 0, the process proceeds to step F3060 shown in FIG. Step F3060 is a process of assigning the designated ALU to the designated host group. Specifically, the MP 2240 registers the designated ALU ID in the specified host group column T1050 of the logical port management table T1000, and further responds with SCSI INQUIRY. ALU LUN for registration is registered, the designated ALU ID is registered in the column T3010 of the bind management table T3000, the bitmap in which all the bits are 0 is registered in the column T3020, and the processing ends.

In this embodiment, ALU LUNs are numbered sequentially from number 243. By the ALU allocation processing of the configuration control unit, the host computer 1000 having the host WWN registered in the designated host group can recognize the ALU via the SAN 4000. On the host computer 1000, the designated ALU is connected to the connection destination. It is possible to form a raw device.

Here, creation of a bind group is explained. As a precondition for creating a bind group, an LDEV to be registered in the created bind group needs to be created in advance. That is, the storage administrator inputs an LDEV creation request specifying an ID indicating a physical storage area and a logical storage capacity to the storage apparatus management program 3110. The storage apparatus management program 3110 uses the LDEV creation API provided by the management server 2500 to transmit the request instruction to the configuration control unit included in the storage apparatus 2000.

Upon receiving the instruction, the configuration control unit registers the ID of the new LDEV in the column T2010 of the logical volume management table T2000, secures a logical storage area with the specified capacity in the specified physical storage area, and creates a logical volume. To do. When creation of the logical volume is processed normally, the configuration control unit registers the capacity specified in the column T2020, registers the ID of the physical storage area specified in the column T2030, and registers the new logical volume in the request source. The LDEV ID is returned and the LDEV creation process is terminated. At this time, the columns L2040 and T2050 of the new LDEV ID are blank.

The storage administrator inputs a bind group creation request specifying the bind group ID and one or more LDEV IDs to be registered in the new bind group to the storage apparatus management program 3110. The storage apparatus management program 3110 uses a bind group creation API provided by the management server 2500. The configuration control unit included in the storage apparatus 2000 receives a bind group creation request from the management server 2500 and starts creation processing.

Specifically, the configuration control unit extracts the bind group ID and the LDEV ID from the request, registers a value indicating the SLU attribute in the designated LDEV ID column T2040 in the logical volume management table T2000, The designated bind group ID is registered in the column T2050, and the process is terminated. The same process is performed when an LDEV is added to the bind group.

FIG. 17 is a diagram illustrating an example of a path from the logical ports 2110j and 2110k to the SLU 2210m in the storage apparatus 2000. Here, there will be described restrictions when there is an existing route in one bind group 2600j and a new route is added to the bind group 2600j. The following description shows a case where the ALU 2210k routed through the new routes 1 to 3 is different from the ALU 2210j routed through the existing route. The ALU consumes the LDEV ID but does not have a logical storage area or a cache memory area. Therefore, since the ALU does not become Single Point Failure, it is not necessary to take a redundant configuration.

The new route 1 is a route that reaches the bind group 2600j via the same logical port 2110j and host group 2111j as the existing route. Therefore, even if a new path 1 is added, the redundancy does not change both physically and logically, load distribution is not performed, and an extra LLU ID of the ALU is consumed. The setting instruction must be rejected.

The new route 2 is a route that reaches the bind group 2600j via the logical port 2110k and the host group 2111k different from the existing route. Here, it is assumed that the host WWN registered in the host group 2111k through which the new route 2 passes and the host WWN registered in the host group 2111j through which the existing route passes are the same. The configuration of the existing route and the new route 2 is a dual port configuration for the host computer 1000 (or virtual server) having the host WWN registered in the host groups 2111j and 2111k, and has the effect of improving the redundancy of the storage area network. Obtainable.

The new route 3 is a route that reaches the bind group 2600j via the logical port 2110k and the host group 2111m different from the existing route. Here, it is assumed that the host WWN registered in the host group 2111m through which the new route 3 passes and the host WWN registered in the host group 2111j through which the existing route passes are different. Since the setting of the new route 3 is to assign the SLU 2210m in the bind group 2600j to the new host computer (or virtual server), when setting the new route 3, the administrator having security authority determines whether or not the setting can be made. There is a need to prompt.

Bind group assignment processing including the above restrictions will be described with reference to FIG. 18, FIG. 19, and FIG. In the bind group assignment process, the storage administrator inputs a bind group assignment request specifying a bind group ID and an ALU ID to the storage apparatus management program 3110. The configuration control unit included in the storage apparatus 2000 receives this allocation request via the management server 2500, and starts and executes bind group allocation processing. In the following description, the case where the bind group 2600j is not assigned to the ALU 2210k in FIG. 17 at the start of the bind group assignment process will be described as an example.

First, the configuration control unit extracts the bind group ID and the ALU ID from the received request (step F4010), and extracts all the host groups to which the designated ALU is assigned from the logical port management table T1000, that is, the column T1050. All the host group IDs in the column T1030 of the row where the designated ALU ID is registered are extracted, and the extracted host group group is set as a set HG1 (step F4020). Next, the number of elements in the set HG1 is calculated. If the number of elements in the HG1 is 0, the designated ALU is not allocated to the host computer 1000 on the SAN 4000. The management computer 3000 is notified of an ALU unallocation notification prompting the user (F4040), and the process is terminated.

When the number of elements of the set HG1 is 1 or more, the configuration control unit refers to the logical volume management table T2000, extracts the existing ALUs to which the designated bind group is assigned, and sets the extracted ALUs as the set A1 ( Step F4050). Next, referring to the logical port management table T1000, a host group to which an ALU belonging to the set A1 is assigned is extracted, and the extracted host group group is set as a set HG2 (step F4060). The host group IDs of the host groups included in the set HG1 and the set HG2 include the PORT ID.

It is confirmed whether the same host group exists in the elements of the set HG2 and the set HG1 (step F4070). If the same host group exists in the set HG2 and the set HG1, it corresponds to the new path 1 described with reference to FIG. The configuration control unit notifies that the SLU overlaps distribution (step F4080), and asks the requesting storage administrator whether the setting is possible. Specifically, the configuration control unit requests an input of whether or not SLU overlap distribution is possible as a request to the management server 2500, and waits for the process in step F4090 until an input is obtained.

When the requesting storage apparatus management program 3110 receives the SLU duplicate distribution permission input request, it displays on the output unit 1600 that it is SLU duplicate distribution and an input form. The storage administrator who is the requester inputs approval / disapproval on the input form. If the permission is granted, the input value x = 1, and if not, the input value x = 0. The storage apparatus management program 3110 responds to the input request from the management server 2500, and the management server 2500 responds with the input value x to the configuration control unit.

If x is not 1 in step F4100, the configuration control unit ends the bind group assignment process. When x is 1 in step F4100, the process proceeds to step F4110. In step F4110, the host WWN registered in the host groups of the set HG1 and the set HG2 is extracted, the host WWN group of the set HG1 is set as a set H1, and the host WWN group of the set HG2 is set as a set H2. Next, it is confirmed whether the set H1 is a subset of the set H2 (step F4120).

Here, when the set H1 becomes a subset of the set H2, since it corresponds to the new route 2 described with reference to FIG. 17, the configuration control unit advances the processing to step F4160. When the set H1 is not a subset of the set H2, since it corresponds to the new path 3 described with reference to FIG. 17, in order to notify the requesting storage administrator to allocate the SLU in the specified bind group to the new host. The management computer 3000 is notified (step F4130), and the management computer 3000 (storage administrator) having security authority is requested to determine whether or not the specified bind group can be assigned to the specified ALU (step F4140).

When the storage administrator having security authority permits the bind group assignment via the storage device management program 3110 of the management computer 3000, the input value y is 1, and when the storage group is not possible, the input value y is 0. If the input value y is 0 in step F4150, the configuration control unit ends the bind group assignment process.

When the input value y is 1 in step F4150, the configuration control unit registers the ID of the designated bind group in the designated ALU column T2050 of the logical volume management table T2000, and designates the designated bind group (for example, ALU 2210k) in the designated bind group (ALU 2210k). For example, bind group 2600j) is assigned (step F4160), and the process is terminated. In step F4160, the bind management table T3000 may be updated.

21 and FIGS. 22 to 24 are diagrams showing examples of flowcharts of processing for binding an SLU to an ALU, that is, setting an SLU to be accessed via the ALU.

FIG. 21 is a diagram showing a flowchart F5000a that is an example of the binding process when the storage administrator designates the SLU and the ALU. The bind process is a process performed by the configuration control unit included in the storage apparatus 2000. The storage administrator transmits a bind request specifying the SLU and ALU to the storage apparatus management program 3110, and the configuration control unit receives the bind request via the management server 2500 and starts the bind process.

In response to the bind request, the configuration control unit extracts the designated SLU ID and ALU ID (step F5010a), refers to the logical volume management table T2000, and extracts the bind group assigned to the specified ALU. That is, the bind group ID of the column T2050 in the row where the designated ALU ID is registered in the column T2010 is extracted, and the extracted bind group group is set as a set B (step F5020a). Further, referring to the logical volume management table T2000, the bind group to which the designated SLU belongs is extracted, and the extracted bind group group is set as a set b1 (step F5030a).

The configuration control unit checks whether any element of the set b1 is included in the set B (step F5040a). If included, the configuration control unit designates the specified SLU with the bitmap in the specified ALU column T3020 of the bind management table T3000. The bit corresponding to is set to 1 (step F5050a), and the process is terminated. If it is not included, it is notified that binding is not possible (step F5060a), and the process ends.

FIGS. 22 to 24 are diagrams showing a flowchart F5000b which is an example of processing when the bind destination ALU is not designated. A flowchart F5000b is processing of the storage apparatus management program 3110 of the management computer 3000, and is executed by the processor 3200. The storage administrator specifies the host computer or virtual server (specified host computer) to which the SLU is assigned, the SLU, and the ALL-ALU flag (any one in the case of 0, all ALUs that can be bound in the case of 1). The bind request is transmitted to the storage apparatus management program 3110 to start processing.

The storage apparatus management program 3110 extracts the host ID and SLU ID (designated SLU ID) of the allocation destination from the request (Step F5010b). The logical volume management table T2000 is acquired from the storage apparatus 2000 via the API provided by the management server 2500, and the bind group to which the designated SLU belongs is extracted, that is, the column T2050 in the row where the designated SLU ID is registered in the column T2010. The bind group is extracted and b2 is set as the extracted bind group (step F5020b).

Then, the storage system management program 3110 refers to the host management table T5000, extracts the host group to which the designated host computer belongs, and sets the extracted host group group as a set H3 (step F5030b). The ALU assigned to the host group of the set H3 is extracted, and the extracted ALU group is set as a set A2 (step F5040b).

Next, for each element a in the set A2, the storage apparatus management program 3110 executes Step F5050b to Step F5090b. In step F5060b, it is confirmed whether or not the bind group b2 is included in the bind group assigned to the element a. If not included, the process proceeds to step F5090b. If included, the element a is registered in the bind list L (step S5060b). F5070b). Next, the value of the designated ALL-ALU flag is confirmed (step F5080b). If the value is 0, the process proceeds to step F5100b. If the value is 1, the process proceeds to step F5090b.

In step F5100b, the storage apparatus management program 3110 confirms the number of elements in the bind list L, and if the number of elements is 0, notifies that there is no ALU that can be bound (step F5130b), and ends the process. When the number of elements is larger than 0, an instruction to bind the designated SLU to the ALU registered in the bind list L is transmitted to the configuration control unit included in the storage apparatus 2000 (step F5110b), and it is confirmed that the bind process has been normally completed. After that, the requester is notified of the ALU in the bind list L (step F5140b), and the process is terminated.

When deallocating an ALU, the configuration control unit included in the storage apparatus 2000 refers to the bind management table T3000, confirms the SLU bind status to the ALU designated to be deregistered, and at least one SLU that is bound. Is present, it rejects the ALU deallocation request. If there is no bound SLU, the configuration control unit deletes the designated ALU from the column T1050 corresponding to the deallocation target host group in the logical port management table T1000.

When deleting an ALU, the configuration control unit of the storage apparatus 2000 refers to the bind management table T3000, confirms the presence of an SLU binding to the ALU designated for deletion, and further refers to the logical port management table T1000. Then, it is confirmed whether the designated ALU is allocated to any of the host groups. The configuration control unit included in the storage apparatus 2000 deletes the designated ALU from the logical volume management table T2000 when there is no SLU binding to the designated ALU and the designated ALU is not allocated to any of the host groups. . The configuration control unit included in the storage apparatus 2000 rejects the ALU deletion when the SLU binding exists or an ALU is assigned to any host group.

When unbinding an SLU, the configuration control unit included in the storage apparatus 2000 extracts the SLU and ALU specified in the unbind request, and sets the value of the bit indicated by the specified ALU and the specified SLU to 0 in the bind management table T3000. To do. If no ALU is specified in the unbind request, all bit values indicated by the specified SLU are set to 0 in the bind management table T3000 regardless of the ID of the column T3010. As a result, access to the designated SLU from the host computer 1000 on the SAN 4000 can be canceled with a single instruction.

As described above, the host computer can access only the SLU bound to the ALU. Thereby, for example, the host computer 1000b belonging to the business B shown in FIG. 1 can access the SLUs 2210f and 2210g, but can be limited so that it cannot access the SLUs 2210c to 2210e. In addition, the virtual server operating on the host computer can be restricted in the same access as the host computer. Since the operation for changing the limitation is not executed unless the permission of the storage administrator is obtained, an erroneous operation or the like is prevented. Furthermore, by allocating the SLU to the virtual server, no extra cost is generated in the control of the logical volume unit.

Since the SLU binding depends on the business application provided in the host computer 1000 and the operation of the Hyper-Visor unit 1400, it may be requested from the host computer 1000 via the SAN, that is, in-band. In the second embodiment, a configuration for performing bind / unbind with In-band will be described.

FIG. 25 is a diagram illustrating an example of the configuration of the storage system according to the second embodiment. A difference from the first embodiment is that the storage device 2000 includes a logical device (CMD: CoMandＭDevice) 2900 for receiving a storage management operation request via In-band. The CMD 2900 is assigned to the logical port 2110n and can be recognized from the host computer 1000n.

In the second embodiment, the local ID in the logical port 2110n of the host group 2111n that provides the CMD 2900 is fixed to FF, but the host group 2111n may be any host group set by the storage administrator. However, since CMD2900 is normally distributed as a LUN, it needs to be separated from the host group to which the ALU belongs. This is because the SCSI command for operating the storage apparatus 2000 is a vendor-unique command, and the Hyper-Visor unit 1400 on the host computer 1000n may not support the vendor-unique command.

FIG. 26 is a diagram illustrating an example of the configuration of the host computer 1000n according to the second embodiment. The host computer 1000n includes a storage management program 1220 in the memory 1200n. The storage management program 1220 has processing for transmitting a vendor-unique SCSI command to the raw device 1310n to which the CMD 2900 is connected. The storage management program 1220 receives an SLU bind request from the business application 1210 or the host administrator via the input unit 1700, converts it into a vendor-unique SCSI command, and transmits the created command to the raw device 1310n.

Upon receiving the command, the CMD 2900 of the storage apparatus 2000 extracts the designated SLU and ALU, requests the configuration control unit included in the storage apparatus 2000 to instruct the bind and unbind processing, and the configuration control unit binds. And unbind processing is executed. The raw device 1310p is a device different from the raw device 1310n, and is the same as the raw device 1310 described with reference to FIG. 3, and the ALU 2210p is the same as the ALU 2210h.

In addition, the storage management program 1220 receives an SLU bind request from the business application 1210 or the host administrator via the input unit 1700, converts it to a vendor-unique SCSI command, and then converts the created command to the raw device 1310p (Example 1). To the raw device 1310). In this case, when the ALU 2010p of the storage apparatus 2000 (ALU 2210h of the first embodiment) receives the command, it extracts the specified SLU and ALU, and binds and unbinds specified to the configuration control unit provided in the storage apparatus 2000. The process is requested, and the configuration control unit executes bind and unbind processes.

The storage management program 1220 in the second embodiment accepts only bind and unbind operation instructions and rejects other operation requests of the storage apparatus 2000. Operations other than bind and unbind should be performed by a storage administrator who is familiar with the operation of the storage apparatus 2000. As a result, operations from a host administrator or business program that has not mastered the operation of the storage apparatus 2000 can be suppressed, and human error or access to unassigned SLUs can be prevented.

As described above, the binding between ALU and SLU can be changed according to the request from the host computer. As a result, it is possible to quickly cope with a change in the configuration of the virtual server on the host computer.

1000: Host computer 2000: Storage device 3000: Management computer 4000: Storage area network 5000: Management network

Claims (15)

1. In a storage system that connects one or multiple host computers and storage devices,
   The storage device
   It comprises a first layer LU (Logical Unit) and a second layer LU that are hierarchically related,
   A configuration controller that distributes the second layer LU to one or more logical groups, and associates the logical group with the first layer LU;
   The first layer LU and the second layer LU receive the designated access request and the second layer LU distributed to the logical group associated with the designated first layer LU. An access control unit for limiting access to be processed in access to the designated second layer LU,
   The host computer
   A storage system that transmits the access request to the storage apparatus.
2. The storage device
   2. The host computer according to claim 1, further comprising a port that is connected to the host computer and that is accessible to the first layer LU, and that only allows access from the registered host computer. The storage system described in.
3. The storage system
   A management computer connected to the host computer and the storage device;
   The storage device
   In response to the correspondence setting request between the first layer LU and the logical group from the management computer, a host computer that can access the first layer LU requested for the correspondence setting is registered in the port. 3. The storage system according to claim 2, wherein if there is not, the correspondence setting request is rejected.
4. The storage system
   A management computer connected to the host computer and the storage device;
   The storage device
   In response to the correspondence setting request between the first layer LU and the logical group from the management computer, the registered port of the host computer that can access the first layer LU requested for the correspondence setting; The correspondence setting request is rejected when the registered ports of the host computers that can access the first layer LU associated with the logical group requested for the correspondence setting are the same. The storage system according to claim 2.
5. The storage device
   If the registered ports are the same, send a setting availability request to the management computer,
   In response to a setting availability response from the management computer to the setting availability request, without rejecting the correspondence setting request,
   The management computer is
   A setting availability request is received from the storage device,
   5. The storage system according to claim 4, wherein the received setting availability request is output, the setting availability response is input, and the input setting availability response is transmitted to the storage apparatus.
6. The storage system
   A management computer connected to the host computer and the storage device;
   The storage device
   In response to the correspondence setting request between the first layer LU and the logical group from the management computer,
   Access to the first layer LU requested for the correspondence setting and access to the host computer registered in the port and the first layer LU associated with the logical group requested for the correspondence setting 3. The storage system according to claim 2, wherein if the host computer registered in the port is different from possible, the correspondence setting request is rejected.
7. The storage device
   If the registered host computer is different, send a setting availability request to the management computer,
   In response to a setting availability response from the management computer to the setting availability request, without rejecting the correspondence setting request,
   The management computer is
   A setting availability request is received from the storage device,
   7. The storage system according to claim 6, wherein the received setting availability request is output, the setting availability response is input, and the input setting availability response is transmitted to the storage apparatus.
8. The storage system
   A management computer connected to the host computer and the storage device;
   The management computer is
   Sending a request to cancel the host computer registered in the port that the first layer LU is accessible,
   The storage device
   Receiving the release request, and rejecting the release request when there is an LU of the second layer distributed to the logical group associated with the LU of the first layer requested to be released. The storage system according to claim 2.
9. The storage system
   A management computer connected to the host computer and the storage device;
   The management computer is
   Send the first layer delete request,
   The storage device
   Receiving the deletion request, and rejecting the deletion request when there is an LU of the second layer distributed to the logical group associated with the LU of the first layer requested to be deleted. The storage system according to claim 2.
10. The storage device
    Comprising a third LU for accepting requests from the host computer;
    Based on the request received by the third LU, assign the second layer LU to the first layer LU;
    The host computer
    The storage system according to claim 1, wherein a request to allocate the second layer LU to the first layer LU is transmitted to the third LU.
11. The storage device
    The storage system according to claim 10, wherein, as a request accepted by the third LU, a request other than an allocation request of the second layer LU to the first layer LU is rejected.
12. In a storage system access control method in which one or a plurality of host computers and a storage device are connected,
    The storage device
    It comprises a first layer LU (Logical Unit) and a second layer LU that are hierarchically related, the LU of the second layer is distributed to one or more logical groups, and the logical group and the first layer LU are distributed. Associate with one LU,
    The host computer
    An access request in which the first layer LU and the second layer LU are specified is transmitted to the storage device, and the storage device
    The access request is received, and access to the second layer LU distributed to the logical group associated with the designated first layer LU is made to the designated second layer LU. An access control method characterized by limiting access to be processed in access.
13. The storage device
    A host computer connected to the host computer and capable of accessing the LU of the first layer is registered, and further includes a port that permits access only from the registered host computer;
    The host computer
    An access request for data designated by the LU of the first layer and the LU of the second layer is transmitted to the storage device.
    When the host computer that has received the access request and has transmitted the access request is a host computer that is registered in the port as being able to access the first layer LU specified in the access request, the access request is 13. The access control method according to claim 12, wherein the access is permitted.
14. The storage device
    Comprising a third LU for accepting requests from the host computer;
    The host computer
    Sending a request to assign the second layer LU to the first layer LU to the third LU;
    The storage device
    13. The access control method according to claim 12, wherein the second layer LU is allocated to the first layer LU based on a request received by the third LU.
15. A storage device connected to one or more host computers,
    A first-layer LU (Logical Unit) and a second-layer LU that are hierarchically associated with each other;
    A configuration controller that distributes the second layer LU to one or more logical groups, and associates the logical group with the first layer LU;
    Upon receiving an access request in which the first layer LU and the second layer LU are designated from the host computer, the first layer distributed to the logical group associated with the designated first layer LU is received. An input / output control unit that limits access to the two-layer LU to be processed in the access to the designated second-layer LU;
    A storage apparatus comprising:

Images