JP2006155658A - Storage device system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem of a conventional disk array not constructed by combining disk devices having different reliability for the disk devices themselves. <P>SOLUTION: A disk array is provided with a drive management means serving as a program identifying the type of a disk device and managing different disk devices separately and a drive management table storing information used by the drive management means. The disk array is also provided with means serving as programs managing an accumulation time of the disk device. These means are a drive lifetime setting means setting a lifetime of a drive, a drive starting/stopping means starting/stopping an ATA disk device according to a plan, and an operation time measurement means measuring an accumulation time of operation. In construction of a RAID, a drive type notification means serving as a program notifying the type of the disk device in construction of the RAID is arranged since awareness about difference in reliability and performance of the disk device is necessary. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention mainly relates to a storage device system control method and a storage device system.

  One type of storage system connected to a computer is a disk array. The disk array is also called RAID (Redundant Arrays of Inexpensive Disks) and is a storage device system having a plurality of disk devices arranged in an array and a control unit for controlling them. In the disk array, read requests (data read requests) and write requests (data write requests) are processed at high speed by the parallel operation of the disk device, and redundancy is added to the data. As disclosed in Non-Patent Document 1, the disk array is classified into five levels according to the type of redundant data to be added and its configuration. The disk device is a storage device having a hard disk, a magneto-optical disk, or the like.

  On the other hand, there are several types of disk devices that use different interfaces for connection with other devices. For example, a disk device (hereinafter referred to as “FC disk device”) having a fiber channel (hereinafter referred to as “FC”) interface that is standardized by ANSI T11, or an ATA that is standardized by ANSI T13. A disk device (hereinafter referred to as “ATA disk device”) having an (AT Attachment) interface.

  ATA disk devices are relatively inexpensive and are used for home desktop computers and the like. On the other hand, FC disk devices are used in business servers that require reliability. For this reason, the FC disk device has higher data input / output (hereinafter referred to as “I / O”) processing performance and higher reliability to withstand continuous operation for 24 hours 365 days than the ATA disk device. .

  When these disk devices are used in a storage device system such as a disk array, the disk devices may be used properly according to the performance, price, etc. required for the disk array. For example, a technology for improving the reliability of data stored in a disk array by using a disk device with higher performance than a disk device for storing data as a disk device for storing parity that is redundant data is patented. It is disclosed in Document 1.

Japanese Patent Laid-Open No. 10-301720 Two authors, David A. Patterson, “A Case for Redundant Arrays of Inexpensive Disks (RAID),” Proceedings of the 1988 ACM Proceedings of the 1988 ACM SIGMOD International Conference on Management of Data), ACM Press, 1988

  In recent years, there has been a demand for physically consolidating a plurality of disk arrays in order to reduce the cost of data management. However, simply consolidating the disk arrays into one physically impairs flexibility as compared to using multiple disk arrays of different types and performance. Therefore, by combining expensive, high-reliability, high-performance disk devices and inexpensive, low-reliability disk devices in a single disk array and using them for each application, the price of the disk array can be reduced and more flexible. There is a growing demand for array construction.

  Conventionally, however, it is difficult to manage the reliability of disk devices in a unified manner, so a storage device system such as a disk array is built by mixing disk devices with different reliability, for example, device life. It was not considered to manage.

  An object of the present invention is to provide a storage device system and a control method for the storage device system that have solved the above problems.

  The present invention has the following configuration. In other words, the storage device system includes a plurality of storage devices having different reliability of the device itself and a control unit that controls the plurality of storage devices. Here, the control unit includes an interface to which the plurality of storage devices are connected, means for detecting the reliability of the storage device itself, and means for managing the detected information. For example, as the detecting means, means for determining the life (usable time) of the storage device by detecting the type of each storage device can be considered. As a management means, a means for managing the usable time (lifetime) of the disk device at a certain point in time by managing the life of the detected storage device and the usage time of the storage device can be considered.

  Furthermore, the storage device system of the present invention may include means for changing the usage pattern of the storage device according to the remaining life of the storage device managed by the management means. For example, if the remaining time of the managed storage device becomes shorter than a predetermined time, there can be a means for using the storage device for data backup acquisition or stopping the use of the storage device. .

  In addition, the storage device system of the present invention may include means for determining in advance the use of the storage device in accordance with the usable time of the storage device managed by the management means. For example, a storage device having a long usable time is distinguished from a storage device having a long usable time as a storage device that normally uses a storage device group having a long usable time, and a storage device that stores backup data in a storage device group having a short usable time A means to do this is conceivable.

  Furthermore, the present invention may include means for controlling these storage devices based on the type of the storage device detected by the detection means. For example, depending on the type of the detected storage device, means for issuing a command specific to this storage device may be provided.

  In the storage device system of the present invention, a drive management table that recognizes the type of storage device and manages different storage devices separately and a drive management table storing management information used by the drive management unit are provided. Also good.

  Furthermore, the storage device system of the present invention may have the following means for managing the accumulated time. For example, there are drive life setting means for setting the life of the drive, drive start / stop means for systematically starting / stopping the ATA disk device, and operation time measuring means for measuring the accumulated time of the operation.

  Further, a configuration having a storage management server for managing an environment in which the above storage device systems are mixed may be employed.

  Further, it may have means for replicating data between storage device systems or within the storage device system, and means for allocating logical storage areas to be subject to data replication based on lifetime management.

The “storage device” includes both a logical storage device and a physical storage device such as a disk device.

  According to the present invention, it is possible to manage disk devices having different reliability, for example, lifespan and life expectancy, in the same disk array. In addition, the user application of the volume can be determined in consideration of the reliability of the disk devices constituting the array. In addition, the user application of the volume can be determined in consideration of the performance of the disk devices that constitute the array. In addition, disk devices with different control methods can be mixed and managed in the same disk array.

  Further, there is an effect that the ATA disk device can be connected to the Fiber Channel arbitrated loop which is an interface for connecting the Fiber Channel disk device.

  Furthermore, it is possible to automatically create an LU in consideration of the performance and reliability of the disk device and to manage the lifetime at the LU level, thereby reducing the burden on the user of managing the storage device system.

              FIG. 1 is a diagram showing a first embodiment of a storage system to which the present invention is applied. A disk array 700, which is an example of a storage device system, is connected to a computer (hereinafter “host”) 100 and a management terminal 500. The interface used for connection with the host 100 or the like may be FC or SCSI. The disk array 700 includes a disk array control unit 200, an FC disk group 310, and an ATA disk group 410.

  The FC disk group 310 includes a plurality of FC disk devices 301. The ATA disk group 410 includes a plurality of ATA disk devices 401. In the present embodiment, the FC disk device and the ATA disk device are exemplified. However, in the application of the present invention, the disk device employed in the disk array is not limited to these disk devices.

  The management terminal 500 includes an input unit 510 such as a keyboard device used by the user when inputting settings related to the disk array 700 and an output unit 520 such as a display device that shows information related to the disk array 700 to the user.

  The disk array control unit 200 includes a CPU 210 that executes a program for controlling the disk array 700, a memory 202, a cache 203 that temporarily stores input / output data from the host 100, the host 100, the disk array control unit 200, and the like. FC I / F 204 that controls transmission / reception of data between them, drive FC I / F 205 that controls transmission / reception of data between the FC disk group 310 and the disk array control unit 200, ATA disk group 410 and disk array control A drive ATA I / F 206 that controls transmission / reception of data to / from the unit 200, and a management I / F 207 that controls transmission / reception of information between the management terminal 500 and the disk array control unit 200. These parts are connected to each other by an internal bus.

  The drive FC I / F 205 is connected to the FC disk group 310 via the FC. In this embodiment, a fiber channel arbitration loop is used as a protocol used for FC. However, the present invention is not limited to this, and other protocols such as FC point-to-point and fabric protocols may be used. good.

  The drive ATA I / F 206 is connected to each ATA disk device included in the ATA disk group 410 via an ATA bus.

  The memory 202 stores various programs that are executed by the CPU 201 to control the disk array 700. Specifically, a RAID control program 210 that controls the operation of the disk array 700 and a management agent 800 that manages the configuration of the disk array 700. The memory 202 also stores various management information. Specifically, a drive management table 240 in which information related to the FC disk group 310 and the ATA disk group 410 is recorded, a logical storage area (hereinafter “FCLU”) 320 created in the FC disk group 310, and an ATA disk group 410 are recorded. The LU management table 245 records information about a logical storage area (hereinafter referred to as “ATA LU”) 420 to be created.

  The RAID control program 210 is further composed of several subprograms. Specifically, the command is issued to issue a command to the FC disk group 310 and the ATA disk group 410, and is executed to manage the FC disk group 310 and the ATA disk group 410. The drive management program 230 is an LU management program 225 that is executed to manage a logical storage area (hereinafter referred to as “LU”) set in each disk group.

  The management agent 800 is further composed of several subprograms. Specifically, in response to an input from the management terminal 500, a drive information setting program 250 that is executed when setting information related to the disk device (hereinafter referred to as “drive information”), and when drive information is output to the management terminal 500 To the array information setting program 270 and the management terminal 500 that are executed when setting information related to the disk array 700 (hereinafter referred to as “array information”) in response to an input from the drive information notification program 260 and the management terminal 500 An array information notification program 280 is executed when outputting array information.

FIG. 2 is a diagram showing an example of the contents of the drive management table 240. The drive management table 240 includes items for registering various information for each disk device included in the disk array 700. Specifically, a “drive No.” column 241 in which the number assigned to the disk device is registered, and the type of the disk device are registered “
“Drive type” column 242, a RAID group (a set of disk devices configuring redundancy) including a disk device, and an “array configuration” column 243 in which the RAID level (hereinafter “array configuration”) is registered are included. “Usage” column 244 for registering the user usage (hereinafter “array usage”) of the RAID group to be registered, “Startup status” column 245 for registering the status of whether the disk device is activated or stopped, The “cumulative time” column 246 in which the cumulative operation time is registered, the “lifetime setting” column 247 in which the life of the disk device is registered, the “lifetime” column 248 in which the life of the disk device is registered, and the life of the disk device Is registered whether or not data is automatically migrated to another disk device (hereinafter referred to as "automatic migration"). There is an automatic migration specification "column 249.

  Specifically, information indicating FC, ATA, and the like is registered in the “drive type” column 242. In the “array configuration” column 243, a number indicating the order in which the RAID groups are created and information indicating the RAID level are registered. For example, a disk device in which “(1) RAID 5” is registered is a disk device included in a RAID group of RAID level 5 created first. In the disk array 700, it is possible to create a plurality of RAID groups, for example, RAID groups of RAID1 and RAID5. Further, the disk devices included in the RAID group may be all or part of the disk devices included in the disk array 700.

  Specifically, information such as a DB indicating a database application and an FS indicating a file system application is registered in the “use” column 244. In the “activation status” column 245, information indicating ON when the disk device is activated and information indicating OFF when the disk device is deactivated are registered. In the “lifetime” column 248, a value indicating the difference between the times registered in the “life setting” column 247 and the “cumulative time” column 246 is stored. In the “automatic transfer designation” column 249, 1 is registered when the data stored in the disk device is set to be automatically transferred, and 0 is set when the data is not set.

  The drive management program 230 is used when the disk array control unit 200 identifies the type of the disk device, measures the operation time of the disk device, controls the start and stop of the disk device, and the disk device reaches the end of its life. This is executed by the CPU 201 when the automatic transition is performed.

  When the disk array control unit 200 sets the type of the disk device in the drive management table 240 according to the input from the management terminal 500, the drive information setting program 250 sets the life of the disk device in the drive management table 240. In addition, the CPU 201 executes the setting when the start and stop states of the disk device are set in the drive management table 240.

  When the disk array control unit 200 notifies the management terminal 500 of the type of the disk device, or notifies the cumulative operation time of the disk device, the drive information notification program 260 indicates the start / stop status of the disk device. This is executed by the CPU 201 when informing, instructing the replacement of the disk device, and when notifying that another disk device has automatically transferred data because the disk device has reached the end of its life.

  The array information setting program 270 is executed by the CPU 201 when the disk array control unit 200 sets the array configuration and sets the array usage according to the input from the management terminal 500.

  The array information notification program 280 is executed by the CPU 201 when the disk array control unit 200 notifies the management terminal 500 of information related to the array configuration and information related to the array usage.

  FIG. 3 is a diagram illustrating an example of a screen displayed on the output unit 520 of the management terminal 500 when a user or administrator of the disk array 700 performs setting of the disk array 700. In this case, the output unit 520 displays the following areas. That is, the drive type setting area 551 for displaying the type of the input disk device, the drive life setting area 552 for displaying the life of the input disk device, and the start / stop conditions of the disk device designated by the user are displayed. A drive start / stop setting area 553, an array configuration setting area 571 for displaying the input array configuration, and an array usage setting area 572 for displaying the input array usage. While viewing this screen, the user or the like inputs necessary data via the input unit and confirms the contents.

  FIG. 4 shows an example of a screen displayed on the output unit 520 when a user or administrator of the disk array 700 acquires information on the disk array 700 via the management terminal 500. In this case, the output unit 520 displays the following areas.

  In other words, a drive type display area 561 that displays the type of the disk device that the disk array 700 has, an operation time display area 562 that displays the cumulative operation time of the disk device that the disk array 700 has, and the start / stop status of the disk device are displayed. A drive status display area 563 for displaying, a replacement instruction display area 564 for displaying a disk device replacement instruction, and an automatic transition notification display area 565 for displaying that data has been automatically migrated to another disk device because a certain disk device has reached the end of its life. An array configuration display area 581 for displaying the set array configuration and an array usage display region 582 for displaying the set array usage. The administrator or the user can check the state of the disk array 700 by displaying this screen on the management terminal 500.

FIG. 5 shows that the disk array control unit 200 has the FCLU 320 and the ATA LU4.
20 is a diagram illustrating an example of the contents of an LU management table 245 used to manage 20; In the LU management table 245, items for registering the following information exist for each LU in order for the disk array control unit 200 to manage each LU.

  In the “LU No.” column 2451, an identifier arbitrarily assigned to each LU is registered. The “host allocation status” column 2452 indicates whether or not an LU is allocated to the host 100 (hereinafter referred to as “used” if allocated, and “unused” if not allocated). “Yes” or “No” information is registered. In the “SCSI ID” column 2453, when an LU is assigned to the host 100, a SCSI ID number assigned to a logical storage device including the LU is registered. In the “LUN” column 2454, when an LU is assigned to the host 100, a SCSI logical unit number necessary for the host 100 to access the LU is registered.

  In the “capacity” column 2455, the storage capacity allocated to the LU is registered. In the “LU type” column 2456, information for indicating a type indicating whether the LU is an FCLU 320 or an ATA LU 420 is registered. In the “LU remaining life” column 2457, the remaining life of the LU is registered. Specifically, the disk array control unit 200 obtains the remaining life of each disk device constituting the LU from the drive management table 240 registered in the “drive number list” column 2458, and sets the smallest value among them. sign up.

  In the “drive number list” column 2458, information indicating the disk devices constituting the LU is displayed as the drive number. It is registered as a list. In the “evaluation value” column 2459, information indicating an evaluation value serving as a reference for LU selection is registered. A method for calculating the evaluation value will be described later. In the case of a used LU, the maximum value that can be registered in this column (+99999 in this example) is registered in the “evaluation value” column 2459.

  Hereinafter, a case where the disk array control unit 200 constructs the drive management table 240, the LU management table 245, and the RAID group in accordance with a user or administrator instruction will be described.

  When power is turned on to the disk array 700, the disk array control unit 200 executes the drive management program 230 to search for the disk devices connected to the drive FC I / F 205 and the drive ATA I / F 206. To do. As an example of the search for the disk device, there is an example in which the ModeSelect command is used. Specifically, first, the disk array control unit 200 issues a ModeSelect command to each disk device. The disk device that has received the ModeSelect command transmits the page information stored in the disk device to the disk array control unit 200. The page information includes information regarding the type of the disk device. From the page information acquired from the disk device, the disk array control unit 200 identifies the type of the searched disk device.

  Further, the disk array control unit 200 executes the drive information setting program 250 and registers the corresponding information in the “drive No.” and “drive type” items of the drive management table 240 for all detected disk devices. To do.

  The type of the disk device may be input by the user using the input unit of the management terminal 500. In this case, the disk array control unit 200 registers information in the drive management table 240 based on the information received from the management terminal 500.

  After detecting or registering the type of the disk device, the disk array control unit 200 sets the array configuration and the array usage based on a user instruction.

  First, when an instruction to acquire information on “drive No.” and “drive type” input by the user using the input unit 510 of the management terminal 500 is received, the disk array control unit 200 executes the drive information notification program 260. The requested information is acquired from the drive management table 240 and transferred to the management terminal 500 via the management I / F 207. The management terminal 500 displays the received information on a portion where “drive No.” and “drive type” are displayed in the drive type display area 561 displayed on the output unit 520. If the user defines the type of disk device, this process may be omitted.

  Thereafter, the user selects an arbitrary disk device based on the information displayed in the drive type display area 561 displayed on the output unit 520 of the management terminal 500, and uses the array configuration setting screen 571 to select the selected disk device. An instruction to construct a RAID group using the apparatus is input from the input unit 510. In addition, the user inputs the usage of the RAID group to be constructed. The disk array control unit 200 executes the array information setting program and registers the array information such as the RAID group and usage received via the management I / F 207 in the “array configuration” and “use” columns of the drive management table 240. To do.

  The user constructs a RAID group based on the type of the disk device displayed on the drive type display screen 561. At this time, an ATA disk device with low reliability and performance is used for an array application that requires high reliability and high performance. For example, it is possible not to set for an array application for a database. Specifically, when “use” is designated by the user based on information on the life of the disk device described later, it is determined whether or not the life of the disk device matches that purpose. Then, a warning is given to the user according to the determination result.

  For example, the management terminal 500 may issue a warning and warning to the user via the output unit 520 when attempting to set the ATA disk device to a highly reliable and high performance array application. Further, when the disk device is displayed on the drive type display screen 561, different symbols may be used so that different types of disk devices can be distinguished.

  In addition, when creating a RAID group for the first time, the life of the disk device is not taken into account, and when the RAID group is recreated, the remaining life of the disk device is added as a criterion for determining a new application setting. .

  Note that when receiving the information for instructing the acquisition of the “array configuration” and “use” information input from the input unit 510 of the management terminal 500 by the user, the disk array control unit 200 executes the array information notification program to execute the management. Information requested from the management terminal 500 via the I / F 207 is retrieved from the drive management table 240, and the retrieved information is transferred. The management terminal 500 that has received the information displays the acquired information in the array configuration display area 581 and the array usage display area 582 of the output unit 520.

  Next, disk device life management performed by the disk array controller 200 will be described. As described above, the disk array control unit 200 executes the drive management program 230 to control the management of the accumulated operation time of the disk device, planned start / stop, and automatic data migration.

  Here, the “lifetime” of a disk device is the time set by the disk array manufacturer based on specifications such as the average failure interval and recommended operating time of the disk device, or the time set by the disk array manufacturer. Refers to the time set by the array administrator. Here, it is defined as “a time serving as a reference for preventive maintenance for preventing a failure caused by long-term operation of the disk device”. Therefore, it is necessary for a disk device that has exceeded the “lifetime” to be replaced with a new disk device by maintenance personnel or the like after migration of stored data.

  In order to perform life management, first, the disk array 700 performs life setting and automatic migration designation for each disk device. This process may be performed together when the disk array control unit 200 sets various tables in accordance with user instructions. Or it may be performed before that. Life setting for each disk device and automatic migration designation are performed as follows.

  When the user receives an instruction to acquire information on “drive No.” and “drive type” input from the input unit 510 of the management terminal 500, the disk array control unit 200 is requested from the management terminal 500 via the management I / F 207. Transfer the information. The management terminal 500 that has received the information displays the information on the received “drive No.” and “drive type” in the drive type display area 561 of the output unit 520.

  The user identifies the type of the disk device based on the information displayed in the drive type display area 561, and sets the life of the disk device and the necessity of automatic migration via the input unit 510 of the management terminal 500. The disk array control unit 200 that has received the information on the life of the disk device and the necessity of automatic migration of the disk device via the management I / F 207 executes the drive information setting program 252 to execute “life setting” in the drive management table 240. ”And“ automatic transfer designation ”are registered in the received information. The information on “life setting” and “automatic migration designation” may be determined by the disk array control unit 200 uniquely based on the type of the disk device and set in the drive management table 240 without designation by the user. . For example, it is conceivable that an FC disk device is set in the drive management table 240 by predicting that the lifetime is long from the beginning.

  FIG. 6 is a flowchart showing a drive life management procedure performed by the disk array controller 200.

  First, the disk array control unit 200 executes the drive management program 230 to start measuring the operation time of each disk device included in the disk array 700 and integrates the result into the “cumulative time” of the drive management table 240. To go. When measuring the operation time, the operation time is measured based on the clock of the disk array 700 (step 1001). Thereafter, the disk array control unit 200 compares the values of “cumulative time” and “life setting” in the drive management table 240 to determine whether or not the cumulative time has reached the set life. The timing of comparison may be any timing asynchronous with other processing (step 1004).

  If the accumulated time has not reached the lifetime, the processing returns to step 1001. If the cumulative time has reached the set life, it is determined whether 1 is registered in the “automatic migration designation” column 249 of the disk device (step 1005). When the automatic migration is not designated (the registered value is 0), the disk array control unit 200 executes the drive information notification program 260 so that the disk device that has reached the set lifetime is set. As a disk device to be replaced, the management terminal 500 is notified via the management I / F 207.

  The management terminal 500 displays the disk device whose accumulated time has reached the end of life in the replacement instruction display area 564 of the output unit 520 as a disk device to be replaced (step 1006). When automatic migration is specified for a disk device whose cumulative time has reached the end of life (the registered value is 1), the disk array control unit 200 executes the drive management program 230 and the RAID control program 210 to thereby calculate the cumulative time. Is transferred to the disk device that is the automatic migration destination. The disk device that is the automatic migration destination is designated in advance by the user or the like. In this case, one disk device may be designated as an automatic migration destination disk device of a plurality of disk devices (step 1010).

  After the data transfer is completed, the disk array control unit 200 notifies the management terminal 500 of information related to the automatic migration destination via the management I / F 270 (step 1011). When the information is notified, the management terminal 500 displays information regarding the disk device that is the automatic transfer destination in the automatic transfer notification display area 565 of the output unit 520. The drive management means 230 clears the “cumulative time” of the drive management table 240 related to the automatically migrated drive (step 1012), and returns to step 1003.

  In this embodiment, automatic transfer or the like is performed for a disk device whose cumulative time has reached the end of its life, but automatic transfer or the like may be performed for a disk device whose remaining life time is below a certain threshold. In this case, the disk array control unit 200 confirms the value registered in the “lifetime” 2408 and identifies a disk device that is below the threshold.

  On the other hand, the disk array control unit 200 controls the start / stop of the operation of the disk device in order to suppress an increase in the accumulated operation time of the disk device. This control is performed independently of the life management shown in FIG.

  The disk array controller 200 starts / stops the disk device based on a preset trigger. The trigger for stopping / starting the disk device can be, for example, a time during which access from the user at night or the like is completely stopped. Alternatively, a case where the life time of a certain disk device becomes shorter than a certain threshold value can be used as a trigger for starting / stopping the disk device.

  Further, the user may directly specify the start / stop of the disk device via the input unit 510 of the management terminal 500. The management terminal 500 sends an instruction related to activation / deactivation of the disk device to the disk array control unit 200 via the management I / F 207. The disk array control unit 200 controls the start / stop of the disk device based on the received instruction. Thereafter, the disk array control unit 200 registers the start / stop status of the disk device in the “startup status” of the drive management table 240. The drive is started / stopped by issuing a Start / StopUnit command to the drive.

  Note that when acquiring the drive activation status and the accumulated operation time, the user instructs the input unit 510 of the management terminal 500 to acquire information on “activation status” and “accumulated time”. The disk array control unit 200 transfers information designated to the management terminal 500 via the management I / F 207. The management terminal 500 displays the received activation status of the disk device and the accumulated operation time of the disk device in the drive status display area 563 and the operation time display area 562 of the output unit 520.

  The user may replace the disk device in advance based on the accumulated operation time or life expectancy time of the displayed drive.

  Next, a case where the disk array control unit 200 issues a command to a disk device having a different interface will be described.

  For example, the ATA disk device 401 cannot process a command with a plurality of tags. Therefore, when the disk array control unit 200 receives a command for instructing data writing or the like from the host 100, the disk array control unit 200 executes the drive command issuance program 220 to determine a command to be issued to the disk device. Specifically, the disk array control unit 200 refers to the “drive type” in the drive management table 230 when issuing a command to the disk device, and determines whether the command is issued to the FC disk device 301 or the ATA disk device 401. In the case of an ATA disk device, a single command is issued without issuing a command with a plurality of tags.

  According to the present embodiment, the reliability of the disk devices themselves, for example, disk devices having different lifetimes can be mixed and managed in the same disk array 700. In addition, it is possible to determine the usage of the LU in consideration of the reliability of the disk devices constituting the RAID group. In addition, disk devices with different control methods can be mixed and managed in the same disk array.

  FIG. 7 is a diagram showing a second embodiment of a computer system to which the present invention is applied. Only differences from the first embodiment will be described below. In FIG. 7, the disk array 700 ′ has an FC drive housing 300 for storing the FC disk group 310 and an ATA drive housing 400 for storing the ATA disk group. The FC disk device 301 stored in the FC drive housing 300 is connected to the drive FC I / F 205 and the ATA drive housing 400 via the port bypass circuit 330 included in the FC drive housing 300. The FC drive housing 300 and the ATA drive housing 400 are connected by an FC loop 600, specifically, a fiber cable.

  The ATA disk device 401 stored in the ATA drive housing 400 cannot be directly connected to the FC loop 600. Therefore, the ATA drive housing 400 has an FC-ATA conversion I / F 430 for converting the FC interface and the ATA interface. Therefore, the ATA disk device 401 is connected to the FC drive housing 300 via the FC-ATA conversion I / F 430 and the FC loop 600.

  FIG. 8 is a diagram illustrating a configuration of the port bypass circuit 330. The port bypass circuit 330 in this figure has five ports. The port bypass circuit 330 has the same number of selectors 331 as the number of ports, thereby implementing loop connection for each port. The selector 331 forms a signal path through the port if the connection is connected to the port, and forms a signal path through an internal circuit that does not pass through the port if the port is not connected.

  FIG. 9 is a diagram showing the configuration of the FC-ATA conversion I / F 430. As shown in FIG. The FC-ATA conversion I / F 430 is for connecting an ATA disk device to the FC I / F. The FC-ATA conversion I / F 430 includes an FC I / F 431 that is an interface with the FC, an ATA I / F 438 that is an interface with the ATA, a memory 441, a CPU 440 that executes a program in the memory 441, and an FC I / F 431. , An ATA I / F 438, a memory 441, and an internal bus 442 for connecting the CPU 440.

  The memory 441 converts an FC command reception program 432 executed when an FC command is received, an FC data transfer program 433 executed when data is transmitted to and received from the FC, and an FC command into an ATA command. A command conversion program 434 executed at the time, a data buffer 435 for buffering data between the FC and the ATA, an ATA command issue program 436 executed at the time of issuing the converted command to the ATA side, and the ATA An ATA data transfer program 437 executed at the time of data transmission / reception and a command buffer 439 for holding commands and performing temporary buffering for regeneration are stored.

  The CPU 440 executes the FC command reception program 432 to receive a command from the FC, executes the FC data transfer program 433, and performs data transfer between the FC I / F 431 and the data buffer 435. Further, the CPU 440 executes the command conversion program 434 to convert the FC command into an ATA command, and executes the ATA command issue program 436 to issue the converted command to the ATA side. The CPU 220 executes the ATA data transfer program 437 to transfer data between the ATA I / F 438 and the data buffer 435.

  In this embodiment, by using the FC-ATA conversion I / F 430, the disk array control unit 200 and the ATA disk device 401 can be connected via the FC. The FC-ATA conversion I / F 430 converts the command received from the disk array control unit 200 'into a command for the ATA disk device. In addition, data transmitted and received between the disk array control unit 200 ′ and the ATA disk device is also converted by the FC-ATA conversion I / F 430.

  According to this embodiment, an ATA disk device can be connected to a Fiber Channel arbitrated loop that is an interface for connecting an FC disk device.

  FIG. 10 is a diagram showing a third embodiment of a computer system to which the present invention is applied. Hereinafter, differences from the first embodiment will be described. In addition to the configuration of the first embodiment, the computer system of this embodiment includes another disk array 701B connected to the disk array 701A, the host computer 100 connected to the disk array 701B, and the disk arrays 701A and 701B for management. It has a storage management server 10000 which is a computer connected via a network 20000. The two disk arrays are connected to each other via a host FC I / F 204 and a remote copy connection 30000 that each disk array has. The remote copy connection 30000 method may be, for example, fiber channel or ESCON, but is not particularly distinguished in this embodiment. Further, the management I / F 207 of each disk array 701 is used for the connection between each disk array 701 and the management network 20000.

  In addition to the configuration of the first embodiment, the disk array 701 of this embodiment newly has an LU pair management table 246 in the memory 202 in the disk array control unit 2001. The RAID control program also includes a program module for controlling data replication (hereinafter referred to as “remote copy”) between the disk array 701B and the disk array 701B connected to the disk array 701A, and data in the disk array 701A or B. A program module for managing replication (hereinafter referred to as “intra-device replication”) has been newly added. Regarding remote copy and intra-device replication implementation methods, synchronous (data replication completion is not reported to the host device until data replication is completed), asynchronous (data replication to the host device without waiting for data replication completion) In this embodiment, no distinction is made.

  FIG. 11 is a diagram illustrating an example of the LU pair management table 246. The disk arrays 701A and 701B are generated by LU pair or intra-device replication that holds the same data between the disk arrays used for remote copy, using the LU pair management table 246 registered in each memory 202. A pair of LUs holding the same data in the disk array 701 (hereinafter referred to as “LU pair”) is managed.

  The LU pair management table 246 includes items of “LU pair No.”, “copy source LU information”, “copy destination LU information”, and “pair status”.

  In the “LU pair No.” column 246a, an identifier arbitrarily assigned to the LU pair is registered. In this embodiment, identifiers are assigned without distinguishing between LU pairs in the disk arrays 701 and in the disk arrays 701.

  The “copy source LU information” item has three sub-items described below. In the “LU No.” column 246b, the LU No. assigned to the LU in which the original data is stored (hereinafter referred to as “copy source LU”) among the LU pairs given a certain identifier. Is registered. The “pair attribute” column 246c includes information for identifying whether the LU pair is an LU pair based on remote copy or an LU pair based on intra-device replication. Specifically, in the case of remote copy, “remote” In the case of intra-device replication, a value indicating “local” is registered. In the “LU type” column 246d, the type of the disk device that stores the copy source LU is registered.

  The “copy destination LU information” item has three sub-items described below. In the “device ID” column 246e, an identifier assigned to the disk array 701 to which the LU paired with the copy source LU (hereinafter referred to as “copy destination LU”) belongs is registered. For example, the identifier assigned to the remote copy destination disk array 701 is registered in the case of remote copy, and the identifier assigned to the disk array 701 having the copy source LU is registered in the case of intra-device replication. In the “LU No.” column 246f, the LU No. assigned to the copy destination LU is displayed. Is registered. In the “LU type” column 246g, the type of the disk device that holds the copy destination LU is registered.

  In the “pair status” column 246h, the current status of the LU pair is registered. For example, the data stored in each LU of the LU pair is synchronized, and the contents of the data stored in each LU match (hereinafter referred to as “Pair state”), or the data stored between LU pairs A value indicating whether the state is not synchronized (hereinafter, “Split state”) is registered.

  The disk array 701 performs the following using an intra-device replication function. For example, an LU pair in the Pair state is changed to the Split state at an arbitrary time. In this way, the data included in the LU pair at an arbitrary time is stored in the copy destination LU (this process is called “take a snapshot”). Thereafter, the host 100 reads the data from the copy destination LU and writes it to another storage device, such as a tape device, so that the data stored in the LU pair at the time when the snapshot is acquired can be backed up. Further, the copy destination LU itself after acquiring the snapshot may be stored as a data backup.

  It is also possible to combine a remote copy LU pair and an intra-device replication LU pair. For example, when remote copy is performed between the disk arrays 701A and 701B, the copy destination LU included in the disk array 701B is set as the copy source LU for intra-device replication executed by the disk array 701B. Then, a copy destination LU that becomes an LU pair for intra-device replication may be created inside the disk array 701B. As described above, the copy destination LU created here can be used for data backup by setting it to the split state.

  The user can set the copy source LU and the copy destination LU (hereinafter “LU pair”) by freely combining the FCLU 320 and the ATA / LU 420. In particular, the user can set the allocation of LU pairs in consideration of the characteristics of each LU via the storage management server 10000 described later. For example, it is possible to allocate a fiber channel LU 320 using a high-performance, high-reliability FC disk device as a copy source LU and an ATA / LU 420 using a low-cost ATA disk device as a copy destination LU. In addition, it is possible to set an LU that has a LU life expectancy and has a disk device with a short life expectancy, not to be assigned, or to be used only as a copy destination for intra-device replication.

  Next, the storage management server 10000 will be described. FIG. 12 is a diagram showing the configuration of the storage management server 10000.

  The storage management server 10000 includes a CPU 11000, a main memory 12000, a storage unit 13000, a management I / F 14000 for connection to the management network 20000, an output unit 15000, and an input unit 16000. These components are connected to each other via a communication path 17000 such as a bus. The storage unit 13000 holds a list of a storage manager 13100 that is a program executed by the CPU 11000, a storage repository 13200 that is an area for storing information collected from the disk array 701, and a list of disk arrays 700 that are managed by the storage manager 13100. The device list 13300, the LU evaluation item table 13400, and the LU remaining life threshold table 13500 for managing the remaining life of the LU are stored.

  The storage repository 13200 periodically collects information stored in the drive management table 240, LU management table 245, and LU pair management table 246 of the disk arrays 701A and B, and creates a copy of these tables. Is done. The device list 13300, the LU evaluation item table 13400, and the LU remaining life threshold table 13500 are created when the storage management server 10000 executes the storage manager 13100 based on a user instruction.

  FIG. 13 is a diagram illustrating an example of the device list 13300. The device list 13300 is used by the storage management server 10000 to confirm the disk array 701 to be managed by itself and to refer to information necessary for communicating with the disk array 701. In the “target device No.” column 13300a, an identifier arbitrarily assigned to the disk array 701 managed by the storage management server 10000 is registered. A unique identifier assigned to the disk array 701 is registered in the “device ID” column 13300b. In the “IP address” column 13300c, information necessary for the storage management server 10000 to communicate with the disk array 701, for example, when an IP network is used for the management network 20000 is allocated to the disk array 700. Information about the IP address is registered.

  FIG. 14 is a diagram illustrating an example of the LU evaluation item table 13400. In the LU evaluation item table 13400, items to be considered when selecting an LU of an LU pair and evaluation contents in the items are registered. When the storage management server 10000 searches for an LU that matches the user's instruction, the storage management server 10000 evaluates each LU using the items registered in the LU evaluation item table 13400, and the evaluation contents specified in the table are used as the LU management table 245. Are registered in the item of the evaluation value 2459 of the duplicate (hereinafter referred to as “LU management table 245 ′”). Thereafter, based on the evaluation result, the storage management server 10000 selects an appropriate LU and presents it to the user.

  In the “evaluation item” column 13400a, an evaluation item at the time of LU designation that can be selected by the user is registered. In the “evaluation target LU management table column” column 13400b, an item in the LU management table 245 to be evaluated as an evaluation item, for example, an LU type 2456 is registered. The “input condition replacement” column 13400c registers whether or not it is necessary to replace the condition input by the user when evaluating each LU registered in the LU management table 245 '. In the “penalty when conditions are met” column 13400d, when the conditions entered by the user match the contents registered in the items of the LU management table 245 ′ to be evaluated, the “evaluation value” column of the LU management table 245 ′ A value to be added to the value 2459 is stored. In the “penalty upon failure of condition” column 13400e, when the condition entered by the user and the contents registered in the column of the LU management table 245 to be evaluated do not match, the “evaluation value” column of the LU management table 245 ′ A value added to the value 2459 is registered. A specific LU evaluation method will be described later.

  FIG. 19 is a diagram showing an example of the LU life expectancy threshold table 13300. The LU remaining life threshold value table 13300 is used when the storage management server 10000 manages the remaining life of the LU included in each disk array 701. Specifically, it has items for registering the following information.

  In the “threshold number” column 13500a, identifiers arbitrarily assigned to a plurality of set threshold values are registered. In the “device ID” column 13500b, an identifier assigned to the disk array 701 to be managed for life expectancy is registered. In the “target LU list” column 13500c, information indicating an LU whose life expectancy is actually managed among the LUs of the disk array 701 that is the management target is registered. Information indicating a threshold corresponding to the identifier is registered in the “lifetime threshold” column 13500d.

  In this embodiment, the storage management server 10000 confirms the remaining life time for an LU registered as a management target, and if the confirmed remaining life time falls below a set threshold, a warning is output from the storage management server 10000 to the user. The In some cases, the user is prompted to make LU migration settings, or an automatic migration instruction is issued to the disk array device. When the importance of the data stored in the LU is low, the life time of the disk device constituting the LU may be used as a judgment criterion instead of using the remaining life time as a judgment standard. However, in this case, the disk device needs to be replaced immediately.

  In this embodiment, the user uses the storage management server 10000 to create the LU of the disk array 701 in which disk devices having different input / output I / Fs are mixed, and parameters such as the performance and reliability of the disk device. Can be considered. Hereinafter, this processing is referred to as LU allocation processing. Unless otherwise specified, the processing described in the following flowchart is processing executed by the storage management server 10000 executing the storage manager 13100.

  FIG. 15 is a flowchart showing the procedure of the LU assignment process 2000. First, the storage management server 10000 determines whether to execute a process of assigning a copy source LU to the host 100 or a process of assigning a copy destination LU in remote copy or in-device replication according to an instruction from the user (step 2100).

  If it is determined in step 2100 that the copy source LU is allocated, the storage management server 10000 performs a host provided LU allocation process. Details of this processing will be described later (step 2200). Thereafter, the storage management server 10000 inquires of the user whether to allocate the copy destination LU. Specifically, a display for inquiring whether to perform processing on the output unit 15000 is output (step 2300).

  When the copy destination LU assignment is instructed by the user in step 2100 and the copy destination LU assignment is instructed by the user in step 2300, the storage management server 10000 performs a copy destination LU assignment process. Details of this process will be described later (step 2400). Thereafter, the storage management server 10000 ends the LU allocation process. If it is instructed in step 2300 that the copy destination LU allocation process is unnecessary, the storage management server 10000 immediately ends the LU allocation process.

  FIG. 16 is a flowchart showing a detailed procedure of the host-provided LU allocation processing executed in step 2200 of FIG. The storage management server 10000 receives conditions (hereinafter “parameters”) required for the copy source LU from the user. Here, the parameters are “items” in the LU evaluation item table 13400 in addition to basic information necessary for LU creation, such as the target device in which the LU is created, the storage capacity of the LU, the host FC I / F, SCSI ID, and LUN. This is information about the evaluation item registered in the column (step 2210).

  Thereafter, the storage management server 10000 performs allocation candidate LU search processing for selecting an appropriate LU as a copy source LU candidate based on the parameters given by the user. Details of this processing will be described later (step 2220). Thereafter, the candidate of the selected LU is shown to the user, and user confirmation is requested (step 2230).

  In step 2230, when the user approves the presented LU candidate as a copy source LU, the storage management server 10000 stores the LU management table 245 stored in the disk array 701 having the approved LU and its duplicate 245 ′. Update. Specifically, first, the storage management server 10000 changes the “host allocation status” field 245b of the approved LU to “present” for the LU management table 245 ′ stored in the storage repository 13200, and sets “SCSIID”. ”Column 245 c and“ LUN ”column 245 d, the parameter values given by the user are registered. Thereafter, the storage management server 10000 instructs the disk array 701 having the approved LU to reflect the updated contents of the LU management table 245 'in the LU management table 245 (step 2240).

  Thereafter, the storage management server 10000 instructs the disk array control unit 200 to execute LU management based on the LU management table 245 in which the update is reflected, and finishes the host-provided LU allocation processing (step). 2250).

  If the user rejects the presented LU in step 2230, the storage management server 10000 stops the host-provided LU allocation processing (step 2260). In order to automate the process, a process of setting the selected LU as a copy source LU may be performed without presenting the selected LU to the user.

  FIG. 17 is a flowchart showing a detailed procedure of the copy destination LU allocation processing shown in step 2400 of FIG.

  The storage management server 10000 receives parameters for selecting a copy destination LU from the user. Here, the parameter is the information described for the copy source LU described above (step 2410).

  Thereafter, the storage management server 10000 performs allocation candidate LU search processing for selecting a candidate that can be a copy destination LU based on the parameters given by the user. Details of this processing will be described later (step 2420). Thereafter, the storage management server 10000 indicates to the user the copy destination LU candidate selected in step 2420 and requests confirmation of the use of the selected LU (step 2430).

In step 2430, when the user approves the presented LU as the copy destination LU, the storage management server 10000 updates the LU pair management table 246 and the LU management table 245 and the respective replicas 246 ′ and 245 ′. In addition, the storage management server 10000 instructs the disk array 701 that stores the selected LU to reflect the updated contents in each table. Specifically, for the LU pair management table 246 ′, a new LU pair entry is created for the selected copy destination LU, and parameters specified by the user and information on the selected copy destination LU are registered. For the LU management table 245 ′, the “host allocation status” field 245b of the selected LU is changed to “present”, and the parameter values are registered in the “SCSI ID” field 245c and the “LUN” field 245d (
Step 2440).

  Thereafter, the storage management server 10000 performs remote copy processing on the disk array control unit 200 of the disk array 701 having the copy source LU paired with the selected copy destination LU based on the updated pair LU management table 246. Alternatively, it is instructed to execute the in-device replication process, and the copy destination LU assignment process is terminated (step 2450).

  If the presented copy destination LU is rejected in step 2430, the storage management server 10000 stops the host-provided LU allocation processing and ends the host-provided LU allocation processing (step 2460). In order to automate the process, a process of setting the selected LU as a copy destination LU may be performed without presenting the selected LU to the user.

  FIG. 18 is a flowchart showing a detailed procedure of the allocation candidate LU search process shown in step 2220 of FIG. 16 and step 2420 of FIG.

  First, the storage management server 10000 extracts an LU that is not used by the host 100 from the LU management table 245 ′ stored in the storage repository 13200. Specifically, it is determined that an LU for which “present” is registered in the “host allocation status” field 245b of the LU management table 245 ′ is used, and the maximum value (+99999 in FIG. 5) is registered as the evaluation value. . It is determined that the LU registered “None” is unused, and the evaluation value is reset to 0 (step 3020).

  Next, the storage management server 10000 gives an evaluation value to all unused LUs extracted in step 3020 for each item registered in the LU evaluation item table 13400. A specific example will be described later (step 3030).

  Thereafter, the storage management server 10000 outputs the LU having the smallest value in the “evaluation value” column 2459 among all the LUs registered in the LU management table 245 ′ as a copy source LU or copy destination LU candidate ( Step 3040).

  The evaluation values registered in the LU management table 245 ′ are updated together when the storage management server 10000 instructs each disk array 701 to update the LU management table 245 (steps 2240 and 2440). The

  Hereinafter, four specific examples of step 3030 will be described.

  As a first example, consider an evaluation when the user designates “type” as “FC”. The storage management server 10000 confirms that the column of the LU management table 245 ′ to be compared with the user-specified value “FC” is the “LU type” column by checking the “evaluation target LU management table” column 13400b. Check. Thereafter, the storage management server 10000 selects one of the extracted unused LUs and confirms the contents of the “LU type” column of the LU management table 245 ′ corresponding to the LU. When the information registered in the “LU type” field is “FC”, the condition matches with the condition specified by the user. Therefore, the storage management server 10000 enters the “penalty when conditions are met” field 13400d corresponding to the “type” item. In accordance with the registered value, 0 is added to the column of the evaluation value 2459 of the LU management table 245 ′ of the unused LU. If the information registered in the “LU type” field is “ATA”, the condition entered by the user does not match, so the storage management server 10000 displays the “penalty upon failure of condition” column 13400e corresponding to the “type” item. In accordance with the registered value, +100 is added to the evaluation value 2459 column of the LU management table 245 ′ of the unused LU.

  The storage management server 10000 determines whether or not the processing described above, that is, the condition specified by the user for each item is met, and each LU management table 245 ′ uses a predetermined value as an evaluation value according to the result. The process of adding to the evaluation value column is repeated until the evaluation of all items described in the LU evaluation item table 13400 is applied to all the extracted unused LUs. As a result of this processing, the value in the “evaluation value” column 2459 of the LU management table 245 ′ for all unused LUs is determined. In this specific example, it is shown that the storage management server 10000 can provide an LU composed of an FC disk device as a candidate when the user selects “FC” as the type of the disk device.

  As a second example, consider an evaluation when a user creates an LU for backup operation. In an operation mode that holds multiple generations of data backups, an LU composed of an inexpensive disk device or an LU composed of a disk device that is difficult to use for normal business operations due to its short lifespan. It is possible to use it. For this case, in this example, an evaluation item “LU operation” as shown in FIG. 14 is defined. In this evaluation item, when the user designates “LU operation” as “backup”, the storage management server 10000 has “LU type” as “ATA” and “LU remaining life time” as “1000 hours or less”. This item evaluates an LU so as to preferentially provide the LU as a candidate. The evaluation procedure is the same as in the first example. With this evaluation item, the storage management server 10000 can present to the user LU candidates that are more suitable for backup use.

  As a third example, consider an evaluation for creating an LU according to the reliability required by the user. In such a case, it is defined that “LU remaining life time” is evaluated in the evaluation item “reliability” as shown in FIG. This evaluation item is set so that when the user designates “reliability” as “high”, the storage management server 10000 preferentially provides LUs whose “LU remaining life time” is “30000 hours or more” as candidates. This is an item for evaluating the LU. Based on this evaluation item, the storage management server 10000 evaluates the remaining life time using a predefined evaluation criterion, and provides an appropriate LU candidate.

  As a fourth example, consider an evaluation for creating an LU according to the number of copy destinations set by the user. In order to hold a plurality of generations of data as described in the second example, there are cases where a large number of copy destinations are created by using a remote copy or in-device replication function. In such a case, it is defined that “LU type” is evaluated in the evaluation item “number of copy destinations” as shown in FIG. This evaluation item is an item for evaluating each LU so that when the user designates a large number of copy destinations, the storage management server 10000 preferentially searches for an inexpensive ATA LU and presents it to the user as a candidate. Based on this evaluation item, the storage management server 10000 evaluates the LU type using a predefined evaluation criterion based on the number of copy destinations designated by the user, and provides an appropriate copy destination LU candidate.

  In this embodiment, the storage management server 10000 realizes life management at the LU level in the disk array 701 in which disk devices having different input / output I / Fs are mixed. Hereinafter, this processing is referred to as LU life management processing. Unless otherwise specified, each processing is performed by the CPU 11000 of the storage management server 10000 executing the storage manager 13100.

  FIG. 20 is a flowchart illustrating a procedure of LU remaining life management processing.

  The storage management server 10000 periodically monitors the LU remaining life using information in the LU management tables 245 ′ of the disk arrays 701 </ b> A and B that are periodically updated in the storage repository 13200. Specifically, the storage management server 10000, for each LU specified in the “device ID” column 13500b and “target LU list” column 13500c of the LU remaining life threshold table 13500, in the LU management table 245 ′ corresponding to each LU. The value registered in the LU remaining life time 245g is compared with the threshold set in each LU (the value registered in the “residual life threshold” column 13500d corresponding to each LU). If a plurality of threshold values can be applied to one LU depending on the setting status in the “monitoring target LU list” field, the threshold value that maximizes the value in the “residual life threshold value” field is applied (step 4010).

  As a result of the comparison, if the remaining life time of each LU is shorter than the value registered in the “residual life threshold” column 13500d, the storage management server 10000 determines that the LU needs to be output with a warning (step 4020). .

  If there is no warning target LU in step 4020, the storage management server 10000 returns to the processing of step 4010 and continues LU life management.

  If there is an LU to be warned in step 4020, the storage management server 10000 identifies a disk device whose remaining life time has fallen below the threshold for the LU that requires warning output, and designates automatic migration for that disk device. Check if it is. Specifically, the “lifetime” of the drive management table 240 ′ is assigned to all the disk devices described in the “drive number list” column 2458 of the LU management table 245 ′ to which the LU whose remaining life time is below the threshold value belongs. ”Column 2408 is checked, the disk device whose remaining life time is below the threshold is specified, and further, the“ automatic migration designation ”column 249 is confirmed (step 4030).

  In step 4030, if automatic migration is specified for all the disk devices constituting the LU whose remaining life time is below the threshold value, the storage management server 10000 outputs only a warning. Here, if the disk array device has not started the automatic migration operation, the automatic migration is instructed. Specific examples of warning contents include “monitoring time”, “disk array device ID”, “LU No.”, “drive number whose remaining life time is below the threshold”, “device ID and LU No. of the associated copy source or copy destination LU. . ".

  Note that the information of the LU pair management table 246 held in the storage repository 13200 is used when outputting the “related copy source or copy destination LU device ID and LU No.”. By outputting the “related copy source or copy destination LU device ID and LU No.”, the user can immediately determine the possibility that the alerted LU will affect other copy source or copy destination LUs. Recognize and, in some cases, manually change LU pairs due to excess life (step 4040).

  In step 4030, if at least one of the disk devices constituting the LU whose remaining lifetime is below the threshold is not designated for automatic migration, the storage management server 10000 outputs a warning to the user and performs migration. Outputs a message prompting specification or drive replacement instructions. As a specific example of the warning content, in addition to the content described in step 4040, “drive number as a migration destination candidate” can be cited.

  When determining the “migration destination candidate drive No.”, the storage management server 10000 has not set the array configuration with the same drive type using the information of the drive management table 240 ′ held in the storage repository 13200. Decide to search for unused disk units. With the output of “migration destination candidate drive number”, the user can reduce the burden of searching for a migration destination LU for designating migration. The disk array may automatically determine the automatic migration destination of the disk device, so that the output of “drive No. as migration destination candidate” may be omitted (step 4050).

  According to this embodiment, when creating an LU of a disk array in which disk drives with different input / output I / Fs and reliability are mixed, the performance and reliability of the disk drive are taken into consideration based on the characteristics specified by the user. LU can be created. If the disk array has remote copy or in-device replication functions, the copy destination LU is automatically determined based on the characteristics specified by the user in consideration of the performance and reliability of the disk drive. Can do. Furthermore, it is possible to manage the life of the disk devices in the disk array that are difficult for the user to be aware of at the LU level, which is a logical volume.

  Note that the storage management server 10000 of this embodiment has an LU creation and LU level life management in consideration of the performance and reliability of the disk device even if the disk array 700 is configured as in the second embodiment. It can be performed.

  Further, the LU unit life management performed by the storage management server 10000 in this embodiment may be executed by the control unit 200 of the disk array 700 in the first embodiment. In this case, the parameter in the life management for each disk device performed in the first embodiment may be changed to the LU remaining life. When an LU whose LU life is below a certain threshold is found, a disk device that constitutes the LU is detected, and a process for confirming the presence or absence of automatic migration or the like may be performed for each of the disk devices.

It is a system configuration figure in a 1st embodiment. It is a figure which shows the drive management table in 1st Embodiment. It is a figure which shows the input part in 1st Embodiment. It is a figure which shows the output part in 1st Embodiment. It is a figure which shows an example of the LU management table in 1st Embodiment. 4 is a flowchart of a drive life management operation in the first embodiment. It is a system configuration figure in a 2nd embodiment. It is a figure which shows the port bypass circuit in 2nd Embodiment. It is a figure which shows FC-ATA conversion I / F in 2nd Embodiment. It is a system configuration figure in a 3rd embodiment. It is a figure which shows an example of the LU pair management table in 3rd Embodiment. It is a block diagram of the storage management server in 3rd Embodiment. It is a figure which shows an example of the apparatus list | wrist in 3rd Embodiment. It is a figure which shows an example of LU evaluation item table in 3rd Embodiment. It is a flowchart of LU allocation processing in the third embodiment. 14 is a flowchart of host LU allocation processing in the third embodiment. 15 is a flowchart of a copy destination LU allocation process in the third embodiment. 14 is a flowchart of allocation candidate LU search processing in the third embodiment. It is a figure which shows an example of the LU life expectancy threshold table in 3rd Embodiment. 14 is a flowchart of LU life management processing in the third embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... Host, 200 ... Disk array control part, 310 ... Fiber channel disk group, 410 ... ATA disk group, 500 ... Management terminal, 700 ... Disk array, 10000 ... Storage management server

Claims (14)

A plurality of disk devices including a plurality of first type disk devices and a plurality of second type disk devices;
A controller connected to the plurality of disk devices,
The predetermined lifetime value for the first type disk device is different from the predetermined lifetime value for the second type disk device,
The control unit holds a predetermined lifetime value and a cumulative operation time value of each of the plurality of disk devices,
The control unit configures a first type RAID group using the plurality of first type disk devices and a second type RAID group using the plurality of second type disk devices. A storage device system. The storage device system according to claim 1,
When the control unit receives information on the use of the RAID group, the control unit outputs a warning based on the received information and a predetermined lifetime value of the disk device constituting the RAID group. system. The storage device system according to claim 1,
The plurality of first type disk devices are configured with a first type logical storage area, and the plurality of second type disk devices are configured with a second type logical storage area. And
The control unit holds, for each logical storage area, a life expectancy based on a predetermined life for a disk device constituting the logical storage area and an accumulated operation time of the disk device. Storage system. The storage device system according to claim 3,
The control unit holds, for each disk device, a lifespan that is a difference between a predetermined lifetime for the disk device and a cumulative operation time of the disk device,
The control unit stores, for each logical storage area, the shortest remaining life time among the remaining life times of the plurality of disk devices constituting the logical storage area as the remaining life time of the logical storage area. Equipment system. The storage device system according to claim 1,
The control unit receives a predetermined life value for each disk device from a management terminal connected to the control unit. The storage device system according to claim 1,
The storage device system characterized in that the predetermined lifetime value for each disk device is a value determined by an administrator of the storage device system. The storage device system according to claim 1,
A storage system characterized in that a predetermined lifetime value for each disk device is a value determined based on a specification of the disk device. The storage device system according to claim 1,
The storage device system, wherein the predetermined lifetime value for each disk device is a value based on a value determined by a manufacturer of the disk device. The storage device system according to claim 1,
A storage system characterized in that a predetermined lifetime value for each disk device is a value determined based on the type of the disk device. The storage device system according to claim 3,
The control unit holds the backup data of data stored in a certain logical storage area in another logical storage area based on the value of the remaining life of each logical storage area. The storage device system according to claim 10,
The storage system is characterized in that the logical storage area in which the backup data is held is selected based on the value of the life expectancy of each logical storage area. The storage device system according to claim 3,
The control unit outputs a value of a life expectancy time of each logical storage area. The storage device system according to claim 3,
The control unit is configured to migrate data stored in a logical storage area to another logical storage area when the value of the life expectancy time of a logical storage area becomes shorter than a predetermined value. Storage system. A storage device system for storing data,
A plurality of storage devices with different reliability;
And a control unit connected to the plurality of storage devices.

Images