JP2018194947A - Storage device, control program and control method - Google Patents

Abstract

To reduce the time of stopping of a virtual tape system due to update of software.SOLUTION: A storage device comprises: a front end part having a cache device 21; a back end part having a memory device 3; a prediction unit 63 which predicts data which will be required of the cache device 21 within a prescribed time, on the basis of a history of access to the cache device 21; a storing unit 63 which performs control to read the predicted data from the memory device 3 and store it into the cache device 21; and an update unit 64 which controls update of software applied to the back end part, in a state where the front end part is operating.SELECTED DRAWING: Figure 8

Description

  The present invention relates to a storage apparatus, a control program, and a control method.

  A virtual tape system having a tape library and a virtual tape device is known. The tape library stores a plurality of physical volumes (hereinafter referred to as PV) such as magnetic tape cartridges and optical disk cartridges.

  A virtual tape device is an example of a storage device provided between a host device and a tape library, and is a TVC (Tape that stores a logical volume (hereinafter referred to as LV) that is a volume that can be recognized by the host device. Volume Cache). The TVC is realized by a storage device having a RAID (Redundant Array of Inexpensive Disks) configuration, for example.

  The virtual tape device includes various components such as a component that controls an IF (Interface) with respect to a host device, a component that controls an IF with respect to a tape library, and a component that controls a TVC, a tape library, and the like. Each of these components may be realized by a server having a processor executing software (SW), for example, firmware (FW). Each component may be made redundant at the hardware (HW) level in order to increase the availability of the virtual tape system.

Japanese Patent Laid-Open No. 9-288571 JP 2007-219696 A JP-T-2006-501535

  In the above-described virtual tape system, when the FW of a component (including a TVC or a tape library) is updated, the worker may perform work in a state where, for example, the business is stopped and all components are stopped. . As an example, when each component is duplicated and two workers perform a firmware upgrade (hereinafter sometimes referred to as “FU” (Firmware Update)) for each system, it takes about 6 hours of work. Downtime may occur.

  As described above, since the components of the virtual tape device are made redundant, it is also possible to stop the other system during the operation of one system and perform the FU of the other system component (FU one by one). Conceivable. However, since the redundancy of the component is in preparation for a failure of the component during normal operation (during business), there is a case where operation in one system is not allowed during the FU.

  In one aspect, an object of the present invention is to shorten a stop period of a virtual tape system due to software update.

  In one aspect, the storage device may include a front end unit having a cache device, a back end unit having a storage device, a prediction unit, a storage unit, and an update unit. The prediction unit may predict data requested by the cache device within a predetermined time based on a history of accesses to the cache device. The storage unit may perform control of reading predicted data from the storage device and storing the read data in the cache device. The update unit may control the update of software applied to the back end unit in a state where the front end unit is operating.

  In one aspect, it is possible to shorten the stop period of the virtual tape system due to software update.

It is a block diagram which shows the structural example of the virtual tape system as an example of one Embodiment. It is a figure which shows the operation example of a virtual tape system. It is a figure which shows an example of the front end part of a virtual tape system, and a back end part. It is a flowchart which shows the operation example of FU of the virtual tape system which concerns on a comparative example. 5 is a flowchart illustrating an operation example of FW update processing of each component illustrated in FIG. 4. It is a figure which shows the relationship between the FU work time which concerns on a comparative example, and business stop time. It is a figure which shows an example of the relationship between FU work time and work stop time which concern on one Embodiment. It is a block diagram which shows the function structural example of the virtual tape system which concerns on one Embodiment. It is a figure which shows an example of LV list information. It is a figure which shows an example of recall historical information. It is a flowchart which shows the operation example of FU of the virtual tape system which concerns on one Embodiment. 12 is a flowchart illustrating an operation example of FW update processing of a VLP (Virtual Library Processor) illustrated in FIG. 11. 12 is a flowchart illustrating an operation example of FW update processing of the back-end unit illustrated in FIG. 11. It is a flowchart which shows the operation example of the preload determination and FU execution determination process shown in FIG. It is a flowchart which shows the operation example of FU of the front end part which concerns on one Embodiment. It is a flowchart which shows the operation example of FU of the front end part which concerns on one Embodiment. It is a block diagram which shows the hardware structural example of the computer which concerns on one Embodiment.

  Embodiments of the present invention will be described below with reference to the drawings. However, the embodiment described below is merely an example, and there is no intention to exclude various modifications and technical applications that are not explicitly described below. For example, the present embodiment can be implemented with various modifications without departing from the spirit of the present embodiment. Note that, in the drawings used in the following embodiments, portions denoted by the same reference numerals represent the same or similar portions unless otherwise specified.

[1] One Embodiment [1-1] Configuration Example of Virtual Tape System FIG. 1 is a block diagram illustrating a configuration example of a virtual tape system 1 as an example of an embodiment, and FIG. It is a figure which shows an example. The virtual tape system 1 is an example of a storage device that provides a virtual tape library to the host device 4. For example, the virtual tape device 2, a plurality of (two in the example of FIG. 1) tape library 3, and an operation A terminal 5 and a monitoring device 6 may be provided. A host device 4 may be connected to the virtual tape device 2.

  The host device 4 is an example of an information processing device that accesses the tape library 3 via the virtual tape device 2. Examples of the host device 4 include a main frame.

  In the host device 4, for example, a VTCP (Virtual Tape Control Program) 41 may be executed. The VTCP 41 performs various controls related to the operation of the virtual tape system 1. The control by the VTCP 41 may include, for example, control related to volume mounting / demounting in the virtual tape device 2.

  The operation terminal 5 is an example of an operation console that provides a user IF for a component to be operated among the components of the virtual tape device 2. The operation terminal 5 may be an information processing apparatus such as a PC (Personal Computer) or a server, for example, and may be connected to the virtual tape device 2 via a LAN (Local Area Network).

  The monitoring device 6 is an example of a control device, and may collect various information and issue instructions for controlling the FU shortening process according to an embodiment. The monitoring device 6 may be an information processing device such as a PC or a server, and may be connected to the virtual tape device 2 via a LAN.

  The tape library 3 is an example of a storage device, stores a plurality of PVs, and accesses the PVs according to instructions from the virtual tape device 2. PV is an example of a recording medium that stores data. Examples of PV include magnetic tapes such as LTO (Linear Tape Open) cartridges, optical disk cartridges, and the like.

  As shown in FIG. 1, the tape library 3 includes, for example, a plurality of (three or more in the example of FIG. 1) drives 31 and one or more (one in the example of FIG. 1) robots 32. It's okay. Each of the drives 31 is an example of a medium processing apparatus that records and reproduces data with respect to the PV, and the robot 32 is an example of a transport apparatus that picks up, conveys, and inserts the medium cartridge into the drive 31. The number of drives 31 and the number of robots 32 are examples, and any number may be provided in the tape library 3.

  The virtual tape device 2 is provided between the host device 4 and the tape library 3, and gives various instructions to the tape library 3 in response to requests from the host device 4.

  As illustrated in FIG. 1, the virtual tape device 2 may include, for example, a TVC 21, an FC switch 22, a VLP 23, a LAN hub 24, an ICP 25, an IDP 26, a PLP 27, and a PCU 28. FC is an abbreviation for Fiber Channel, and VLP is an abbreviation for Virtual Library Processor. ICP is an abbreviation for Integrated Channel Processor, IDP is an abbreviation for Integrated Device Processor, PLP is an abbreviation for Physical Library Processor, and PCU is an abbreviation for Power Control Unit.

  Any of these components indicated by reference numerals 22 to 28 may be made redundant. For example, a component with “# 0” or “# 2” added to the block becomes “0 series” or “# 0 group” together with the tape library # 0. Also, the components with “# 1” or “# 3” added to the block are “1 system” or “# 1 group” together with the tape library # 1.

  In normal operation, both the # 0 group and the # 1 group may operate. When one of the components fails, the component is in a one-system operation.

  The TVC 21 is an example of a cache device, and stores data of the LV 211 (see FIG. 2) related to access between the host device 4 and the tape library 3.

  When the TVC 21 stores the data of the LV 211, the LV 211 is in an “on-cache” state. When the LV 211 related to the access from the host device 4 is on-cache, the virtual tape device 2 may perform processing related to the access to the on-cache LV 211 on the TVC 21 and return a response to the host device 4.

  On the other hand, the LV 211 that does not exist in the TVC 21 is in an “off cache” state. When the LV 211 related to the access from the host device 4 is off-cache, the virtual tape device 2 may read the data related to the LV 211 from the PV of the tape library 3 onto the TVC 21. As described above, after the off-cache LV 211 is set to the on-cache, the virtual tape device 2 may perform processing related to access to the LV 211 on the TVC 21 and return a response to the host device 4. Reading the data of the off-cache LV 211 from the PV of the tape library 3 to the TVC 21 is called “recall”.

  The TVC 21 may include a plurality of storage devices, and a RAID may be configured by these storage devices. Examples of the storage device include a magnetic disk device such as an HDD (Hard Disk Drive) or a semiconductor drive device such as an SSD (Solid State Drive). The TVC 21 may provide a file system using a disk group such as RAID.

  The FC switch 22 switches communication paths between devices in a network such as a SAN (Storage Area Network) connected via an FC cable or the like, and connects these devices so that they can communicate with each other. In communication via FC, for example, data transfer such as LV211 or PV is performed.

  The section connected by FC is a section indicated by a solid line in the example of FIG. For example, FC connections are made between the TVC 21 and the FC switch 22, between the FC switches 22, and between the FC switch 22 and the VLP 23. The FC switch 22 and the ICP 25, the FC switch 22 and the IDP 26, the ICP 25 and the host device 4, the IDP 26 and the drive 31, and the PLP 27 and the robot 32 may be FC-connected.

  The VLP 23 is an example of a control device, and is connected to the host device 4 via a control IF (see FIG. 2) by a LAN, communicates with the host device 4 (for example, VTCP 41), resource management in the virtual tape device 2, and The FU shortening process according to an embodiment is controlled.

  Resource management by the VLP 23 may include management of the LV 211 and PV, control of mounting and unmounting, and the like.

  Note that one of the VLPs 23 (for example, the 0-system VLP 23) may operate as a master and the other (for example, the 1-system VLP 23) may wait as a slave. In this case, when an abnormality occurs in the master VLP 23, the slave VLP 23 is promoted to the master and can take over the processing of the VLP 23.

  The LAN hub 24 switches communication paths between devices in a LAN connected via a LAN cable or the like, and can connect these devices so that they can communicate with each other. In communication via a LAN, for example, control information such as commands and data is transferred.

  Note that the section connected by the LAN is a section indicated by a broken line in the example of FIG. For example, each of the VLP 23, the ICP 25, the IDP 26, and the PLP 27 may be LAN-connected to the LAN hub 24. The LAN hub 24 may be connected to the host device 4 (VTCP 41), the operation terminal 5, and the monitoring device 6 through a LAN. Each of the TVC 21 and the FC switch 22 may be connected to the LAN hub 24 via a LAN.

  The ICP 25 controls the host IF with the host device 4 and performs emulation and compression processing of the virtual drive 251 (see FIG. 2). For example, the ICP 25 may emulate the host device 4 with the LV 211 as the virtual drive 251, compress the write data for the virtual drive 251, and store the compressed data in the LV 211.

  The ICP 25 is equipped with an HBA (Host Bus Adapter) that communicates with the host device 4 and is connected to the host device 4 using a storage connection interface such as FC, FCLINK (registered trademark), or OCLINK (registered trademark). Good. As an HBA, an interface card compliant with FC or the like can be cited.

  The IDP 26 may perform various controls for the tape library 3 via an actual drive IF such as FC (see FIG. 2) in response to an instruction from the VLP 23. The control by the IDP 26 may include, for example, control for the drive 31 of the tape library 3.

  The PLP 27 may include a controller 271, and the controller 271 may control the robot 32 of the tape library 3. The controller 271 may be software for controlling the robot 32, for example. The control by the controller 271 may include control of the conveyance of the PV by the robot 32, mounting to the drive 31, and unmounting.

  The PCU 28 performs power control of the virtual tape device 2. For example, the PCU 28 may perform power control of each component of the virtual tape device 2 via the LAN hub 24 as indicated by a one-dot chain line in FIG.

  The components of the virtual tape system 1 described above may be categorized into a front end unit 11 indicated by hatching in FIG. 3 and a back end unit 12 indicated by hatching.

  The front end unit 11 is an example of a component group that performs processing related to the on-cache LV 211 (for example, processing according to access from the host device 4). For example, the front end unit 11 may include a TVC 21 and an ICP 25.

  The back-end unit 12 is an example of a component group that performs processing related to the off-cache LV 211 (for example, processing according to access from the host device 4). The back end unit 12 may include, for example, a VLP 23, an IDP 26, a PLP 27, and the tape library 3. The VLP 23 performs control when the LV 211 related to access from the host device 4 is both on-cache and off-cache, but is assumed to belong to the back-end unit 12 for convenience.

[1-2] Firmware Upgrade of Virtual Tape System Next, firmware upgrade (FU) of the virtual tape system 1 will be described using a comparative example. FIG. 4 is a flowchart showing an example of the FU operation of the virtual tape system 1 according to the comparative example, and FIG. 5 is a flowchart showing an example of the FW update process of each component shown in FIG.

  As shown in FIG. 4, the worker adjusts the schedule of the FU work with the user or administrator (hereinafter referred to as “user”) of the virtual tape system 1 (step S101). The schedule includes, for example, the date and time of the FU work and the work period, in other words, the stop date and time and the stop period of the virtual tape system 1.

  Next, the worker confirms the stop of the virtual tape device 2 (step S102). The stop confirmation may include, for example, confirmation of the presence / absence of I / O (Input / Output) from the host device 4 in the virtual tape device 2. If there is an I / O (Yes in steps S103 and S103), the virtual tape device 2 is not stopped, so the process proceeds to step S102, and steps S102 and S103 are executed until there is no I / O.

  If there is no I / O (No in step S103), the virtual tape device 2 is stopped, so the worker performs FW update processing for each component of the virtual tape system 1 (steps S104 to S107). When the FW update process for all the components is completed, the FU operation ends. In addition, the process of step S104-S107 assumes the case where two workers work in parallel. In the example of FIG. 4, the FU work of the front end unit 11 by the worker A and the FU work of the back end unit 12 by the worker B are performed in parallel.

  Next, the FW update process of each component in steps S104 to S107 will be described. As shown in FIG. 5, the worker connects the work terminal to the FW update target component or the LAN hub 24 (step S111).

  The worker logs in to the FU update target component from the work terminal (step S112), instructs the component to read the updated version of the FW (step S113), and restarts the component (step S114).

  Further, the worker confirms the log related to the application of the updated version of FW (step S115), and if there is no abnormality, logs out from the component (step S116), disconnects the work terminal (step S117), The FW update process ends.

  In the method according to the comparative example described above, in the FU work of the virtual tape system 1, the worker stops the work of the virtual tape system 1 without separating the processes of the front end unit 11 and the back end unit 12, FU components at once. This is because the back end unit 12 may operate when data is manipulated via the front end unit 11. For example, when the host device 4 designates data that is not on the TVC 21 of the front end unit 11, the virtual tape device 2 operates the back end unit 12 to extract data from the tape library 3, and the TVC 21 of the front end unit 11. The process of arranging above is performed.

  Therefore, as illustrated in FIG. 6, the virtual tape system 1 is in a business stop state until the FU work of the front end unit 11 and the FU work of the back end unit 12 are completed. As an example, 20 minutes for preliminary work, 110 minutes for FU work of the front end unit 11 (parallel work of two workers), and 110 minutes of time for FU work of the back end part 12 (parallel work of two workers) , It will be a total of 240 minutes (4 hours) business stoppage time.

  Therefore, in one embodiment, the following method is used to shorten the operation stop time for stopping the operation of the virtual tape system 1 for the FU.

  For example, the virtual tape system 1 according to an embodiment predicts data that is required to be recalled during FU work during business (operation), and places the data on the cache of the front end unit 11. Further, the virtual tape system 1 adjusts the FU available time during which the FU for stopping the business can be executed.

  Thereby, the front end part 11 and the back end part 12 can be separated, and the FU can be alternately performed on the front end part 11 and the back end part 12.

  Therefore, as shown in FIG. 7, for example, the FU work of the back-end unit 12 can be completed during the work, and the FU work of the front-end unit 11 is performed during the period when the virtual tape system 1 is in the work stop state. The period can be limited.

[1-3] Functional Configuration Example of Virtual Tape System Next, a functional configuration example related to FU control of the virtual tape system 1 will be described with reference to FIG.

  In the following description, the on-cache or off-cache LV managed by the TVC 21 is denoted as LV70, and the data storage PV provided in the tape library 3 and storing the LV70 data is denoted as PV80.

[1-3-1] Example of Functional Configuration of Monitoring Device The monitoring device 6 extracts components that can be FU during work, and notifies a worker (hereinafter sometimes referred to as “operator”), Various processes and controls related to the FU of the virtual tape system 1 are performed.

  As illustrated in FIG. 8, the monitoring device 6 may include, for example, a memory unit 61, an information acquisition unit 62, a preload control unit 63, and an FW update unit 64 as one function of the state investigation tool. The memory unit 61 may be realized by a memory provided in the monitoring device 6. Moreover, the information acquisition part 62, the preload control part 63, and the FW update part 64 may be implement | achieved when a control program is run by CPU with which the monitoring apparatus 6 is equipped.

  The memory unit 61 has a storage area for storing information such as LV list information 611 and recall history information 612. Information stored in the memory unit 61 will be described later.

  When the FU of the virtual tape system 1 is performed, the information acquisition unit 62 acquires information used for the FU from the virtual tape device 2 when the component of the back end unit 12 is included as an FW update target.

  For example, the information acquisition unit 62 may acquire information related to the LV list information 611 from the VLP 23 and acquire the host access time from the LV list information 611. The LV list information 611 is information indicating a list of LVs 70 managed by the virtual tape device 2.

  Information related to the LV list information 611 may be acquired from the VLP 23 by executing a vlmcmd query command while logged in to the VLP 23, for example. Thereby, the information of LV70 managed by VLP23 is acquirable.

  FIG. 9 shows an example of the LV list information 611. As shown in FIG. 9, the LV list information 611 includes, for example, “volume name”, “volume group name”, “size” indicating volume capacity, “last access date”, “status”, and “flag”. May be included. The last access date / time indicates the last host access time from the host device 4. In the LV list information 611, the LV 70 in which “M” is set in the flag indicates that the recall is being performed.

  The LV list information 611 is used for monitoring the status of the LV 70 by the monitoring device 6, for example, confirming the presence / absence of the recalled LV 70, monitoring the status of the recalled LV 70, for example, confirming the previous recall history, etc. May be used.

  For example, the information acquisition unit 62 may acquire information of the recall history information 612 based on the LV list information 611.

  FIG. 10 shows an example of the recall history information 612. As illustrated in FIG. 10, the recall history information 612 may include, for example, items of “volume name”, “recall start time”, and “recall end time”.

  As an example, the information acquisition unit 62 monitors whether the LV 70 is being recalled for each entry (LV 70) of the LV list information 611, and collects the recall history of the LV 70 in which “M” is set in the flag. Good. The recall history may be collected by the following procedures (i) to (iv).

  (I) The information acquisition unit 62 issues a vlmcmd query command on the VLP 23 every predetermined period (for example, 60 seconds), and “M” is set in the flag for each LV 70 or is not set (flag is not set) To monitor.

  (Ii) The information acquisition unit 62 monitors whether the state of the flag has changed.

  (Iii) When the flag state changes from “not set” to “M”, the information acquisition unit 62 acquires the row of the LV 70, sets the volume name in the recall history information 612, and sets the current time as the recall start time. Add an entry with set.

  (Iv) When the flag state changes from “M” to “no setting”, the information acquisition unit 62 acquires the row of the LV 70, and the current time at the recall end time for the corresponding entry of the recall history information 612. Set.

  Through the above processing, the recall history for each LV 70 is recorded in the recall history information 612. In the recall history information 612, a plurality of recall histories for one LV 70 may be accumulated.

  The preload control unit 63 controls the preload of the off-cache LV 70 to the TVC 21 based on the information acquired by the information acquisition unit 62 in the FU control. The control may include a preload instruction specifying the LV 70 for the VLP 23. Note that “preload” means that processing for reading off-cache LV 70 data from the PV 80 of the tape library 3 to the TVC 21 is performed before host access. Although the preload is executed at a different timing, the preload may be a process similar to the recall (similar).

  As an example, the preload control unit 63 records the last access time of the LV list information 611, compares the recorded last access time with the current time, and extracts a candidate LV 70 to be recalled within a predetermined time from the current time. Good.

  As another example, the preload control unit 63 records the recall start time of the recall history information 612, compares the recorded recall start time with the current time, and extracts a candidate LV 70 to be recalled within a predetermined time from the current time. May be.

  The “predetermined time” is a time related to the FU work time. For example, 20 minutes for the pre-work, 110 minutes for the FU work of the front end unit 11, and 110 minutes for the FU work of the back end unit 12. The total time of 240 minutes may be added.

  Alternatively, when the FU work of the front end unit 11 is performed after a FU work of the back end unit 12 (for example, at a later date), the “predetermined time” is a time related to the FU work time of the back end unit 12. There may be. For example, the “predetermined time” may be a time of 130 minutes that is a sum of 20 minutes of the preliminary work and 110 minutes of the FU work of the front end unit 11.

  Here, the virtual tape system 1 may be used as a backup or archive destination, and access to the virtual tape device 2 may be scheduled at a predetermined cycle, for example. When access to the virtual tape device 2 is scheduled, the access timing has regularity such as a specific time every day, a specific time of a specific week or day, a specific time of a specific day of the month, etc. Become.

  Therefore, the preload control unit 63 pays attention to the regularity in the access timing for each LV 70 to the virtual tape device 2, and based on the LV list information 611 or the recall history information 612, the current time and the “specific time” And the candidate LV 70 is extracted.

  The preload control unit 63 may sequentially select candidate LVs 70 whose time part of the last access time or recall start time is closest to the current time from the extracted candidate LVs 70, and may control the execution of preloading for the selected candidate LVs 70. .

  When the preload control unit 63 completes the preloading for all the extracted candidate LVs 70, the preload control unit 63 compares the current time with the “specific time” again based on the LV list information 611 or the recall history information 612, and the candidate LV 70 May be extracted. This is because there can be an LV 70 in which a recall occurs within a predetermined time from the current time after completion of preloading.

  Further, the preload control unit 63 may calculate the time from the current time until the next recall of the LV 70 occurs when the preloading of all candidate LVs 70 is completed. This time means a time when no recall occurs, that is, a time during which the front end unit 11 and the back end unit 12 can be separated (updatable time).

  The preload control unit 63 may notify the operator of the calculated time as the FU executable time. Note that the notification may include screen output to a display unit such as a monitor or message transmission by e-mail.

  When the operator performs the FU work of the back-end unit 12 or the front-end unit 11, the possibility of the recall being executed in the FU time zone is considered based on the FU workable time and taking into account a margin such as FU work delay. it can. For example, when it is determined that the recall is unlikely to be executed during the FU operation, the operator may determine to execute the FU and perform the FU operation using the FW update unit 64. The above determination by the operator may be performed by the preload control unit 63 based on conditions such as a margin and a threshold value.

  There may be a case where there is not enough history in the LV list information 611 or the recall history information 612 to determine the candidate LV 70, such as the first day of introduction of the virtual tape system 1. In this case, the operator may determine that the FU operation is not performed (postponed to a later date), or that the FU operation is performed by a method such as the comparative example shown in FIGS.

  Preloading by the preload control unit 63 can suppress the occurrence of recall processing of the LV 70 during FU work. When the recall does not occur, the components of the back end unit 12 such as the IDP 26, the PLP 27, and the tape library 3 do not depend on the processing related to the access from the host device 4, so the back end unit 12 is in business. In addition, FU is possible internally.

  Hereinafter, the process of the preload control unit 63 will be described with an example. As described above, the preload control unit 63 finds the tendency of the recall execution from the host access history of the off-cache LV 70 in order to determine the possibility of the recall execution.

  For example, it is assumed that there are four off-cache LVs 70 of “AAAAAA”, “BBBBBB”, “CCCCCC”, and “DDDDDD”, and the current date and time is 8/31 (Wed) am 9:00. Further, it is assumed that each LV 70 has the following tendency from the LV list information 611 or the recall history information 612.

“AAAAAA” is accessed around pm 9:30 every day.
“BBBBBB” is accessed around am9: 30 every day.
“CCCCCC” is accessed near am 9:30 on an even day.
“DDDDDD” is accessed every Tuesday at around 9:30 am.

  In this case, the preload control unit 63 determines that the recall of “AAAAAA”, “CCCCCC”, and “DDDDDD” is unlikely to be executed within a predetermined time at the current date and time, and “BBBBBB” is recalled. Judging that there is a high possibility of being.

  Therefore, the preload control unit 63 performs preloading in advance and sets the on-cache so that “BBBBBB” is not recalled during the FU operation.

  When a certain LV 70 is set to on-cache and the TVC 21 is full, another LV 70 that is on-cache is evicted from the TVC 21. The cache eviction logic of the LV 70 may be logic based on an LRU (Least Recently Used) method. Alternatively, it may be a logic that selects an LV 70 having a low possibility of host access near the current date and time from the host access history of the on-cache LV 70 and turns it off-cache. Hereinafter, an example of the latter logic will be described.

  The preload control unit 63 searches the oldest file in the access history from the LV list information 611 or the recall history information 612. When the size of the searched file is larger than the file size of the file to be preloaded, the preload control unit 63 caches out the searched file (off-caches).

  When the size of the searched file is smaller than the file size of the file to be preloaded, the preload control unit 63 searches for the second oldest file. When the sum of the sizes of the first and second searched files is larger than the file size of the file to be preloaded, the preload control unit 63 caches out the first and second searched files.

  Similarly, the preload control unit 63 repeats the above processing until the total size of the old files in the retrieved recall history becomes larger than the size of the file to be preloaded, and caches the retrieved file.

  Note that the preload control unit 63 may check the current state of the LV 70 (on-cache or off-cache) as a material for determining whether or not there is a possibility that the recall is executed in the LV 70. If there is no off-cache LV 70, there is a high possibility that the components of the backend will not operate for the time being, and it is possible to stop these and perform FU. Therefore, when there is no off-cache LV 70, the comparison of the current time with the access history and the execution of preload may be suppressed, and the processing of the FW update unit 64 may be shifted.

  As described above, the FW update process of the back end unit 12 by the FW update unit 64 can be executed by preparing a situation in which the execution of recall within a predetermined time from the current date and time is suppressed by the preload control unit 63. Become.

  As described above, the preload control unit 63 is an example of a prediction unit that predicts data requested by the TVC 21 within a predetermined time based on a history of access to the TVC 21. The preload control unit 63 is an example of a storage unit that performs control of reading predicted data from the tape library 3 and storing it in the TVC 21.

  When the preloading by the preload control unit 63 is completed and the FU work is performed (for example, when the operator decides to perform the FU work based on the FU executable time), the FW update unit 64 performs the FW of each component of the backend unit 12. Update processing is performed. The FW update unit 64 may execute the FW update process of the VLP 23 among the components of the back end unit 12 before the process by the information acquisition unit 62 and the preload control unit 63. As described above, the monitoring device 6 is for collecting various information, controlling FU work, and the like via the VLP 23.

  In the FW update process related to each component of the back end unit 12, the following processes (I) to (III) may be performed.

  (I) The FW update unit 64 individually sets the maintenance mode during the update process of the FWs of the VLP 23, the IDP 26, the PLP 27, and the tape library 3 so that the operator can confirm the message at the time of restart. . During the maintenance mode, when a component is restarted, detection of restart in other components is suppressed.

  (II) The FW update unit 64 performs the FW update process of the standby VLP 23 (for example, VLP # 1) in the FW update process of the VLP 23, performs Failover after completing the FW update of the VLP # 1, and performs VLP # 1 is changed to the active system (for example, master) VLP23. Next, the FW update unit 64 performs FW update processing for the VLP 23 (for example, VLP # 0) that has been changed to the standby system, and after completing the FW update for VLP # 0, performs the Failback to change the VLP # 0 to the active system (for example, Master) Change to VLP23. As described above, in the FW update process of the VLP 23, the use of the failover function that is restricted in operation in the comparative example prevents the operation from being stopped and suppresses the decrease in availability to the minimum period. FW update processing can be executed.

  (III) The FW update unit 64 performs a physical library / physical drive control process, for example, PLS (Physical Library Service) / PDS (Physical Device Service) during the FW update process of the IDP 26, PLP 27, and tape library 3. Stop. The PLS is a service that performs IF control with the tape library 3 and requests the PLP 27 for a mount request or the like, and is executed by the VLP 23. The PDS is a service that stores data from the PV 80 to the TVC 21 and saves data from the TVC 21 to the PV 80 in accordance with an access request from the VLP 23, and is executed by the IDP 26.

  Note that at least a part of the processing of the FW update unit 64 described above may be realized by an operator's operation via the monitoring device 6, or the CPU executes a program in which a code for performing the processing described above is written. May be realized.

  As described above, the FW update unit 64 is an example of an update unit that controls the update of software applied to the back end unit 12 in a state where the front end unit 11 is operating.

[1-3-2] VLP Functional Configuration Example As shown in FIG. 8, the VLP 23 may include, for example, an LV information transmission unit 231 and a preload execution unit 232. The LV information transmission unit 231 and the preload execution unit 232 may be realized by executing a control program by a CPU provided in the VLP 23.

  In response to the vlmcmd query command from the information acquisition unit 62, the LV information transmission unit 231 transmits (responds) information on the LV 70 managed by the VLP 23 to the information acquisition unit 62.

  When the preload execution unit 232 receives a preload instruction from the monitoring device 6, the preload execution unit 232 may cooperate with the IDP 26 and the PLP 27 to perform control of reading the data of the designated off-cache LV 70 from the PV 80 of the tape library 3 to the TVC 21 ( (See FIG. 8). The preload execution unit 232 may transmit a preload completion response to the monitoring device 6 when the processing of the IDP 26 and the PLP 27 is completed.

  Note that when the above-described prediction by the monitoring device 6 is lost and a recall request is generated during the FW update process of the back-end unit 12, the component of the front-end unit 11 performs a process related to the request to the request source. A response such as that the device is in use may be sent. As a result, the request source can be prevented from waiting for a response.

[1-4] Operation Example of Virtual Tape System Next, an operation example of FU control in the virtual tape system 1 configured as described above will be described with reference to FIGS.

  As a premise, it is assumed that a monitoring device 6 is connected to the virtual tape device 2 and a state investigation tool is stored in the monitoring device 6 in order to search for FU-capable components.

  Further, when an operator (or a worker) performs FW update processing, the developer requests the operator to update FW. When the operator requests the FW update from the developer, the operator checks whether the FW update of the back-end unit 12 is included. When the requested FW update includes the FW update of the back-end unit 12, the operator sets the FU available time display mode to ON for the monitoring device 6. Based on the above, an example of FU control operation will be described below.

[1-4-1] Operation Example of FU Control of Back End Unit As illustrated in FIG. 11, the monitoring device 6 determines whether or not the FU available time display mode is ON (step S1). If it is not ON (No in step S1), the process ends.

  When the FU available time display mode is ON (Yes in step S1), the monitoring device 6 determines whether the FW of the VLP 23 is the latest (step S2). In the latest case (Yes in step S2), the process proceeds to step S4.

  On the other hand, when it is not the latest (No in step S2), the FW update unit 64 of the monitoring device 6 performs the FW update process of the VLP 23 (step S3), and the process proceeds to step S4.

  In step S4, the monitoring device 6 determines whether or not the FW of the backend unit 12 is the latest, and if it is the latest (Yes in step S4), the process ends. On the other hand, when it is not the latest (No in step S4), the monitoring device 6 performs the FW update process of the back-end unit 12 (step S5), and the process ends.

  When the process illustrated in FIG. 11 ends, the FW update process of the front end unit 11 illustrated in FIG. 15 or FIG. 16 is executed continuously or after a certain time (for example, at a later date).

[1-4-2] Operation Example of VLP FW Update Processing Next, an operation example of the FW update processing of the VLP 23 shown in step S3 of FIG. 11 will be described with reference to FIG.

  As illustrated in FIG. 12, the FW update unit 64 sets both the maintenance modes of VLP # 0 and # 1 to ON (step S11), and logs in to VLP # 1 (standby system) (step S12).

  Next, the FW update unit 64 starts the FU of VLP # 1 (step S13) and waits for the completion of the FU (No in steps S14, S15, and S15). When the FU of VLP # 1 is completed (Yes in step S15), the FW update unit 64 sets VLP # 1 to the master VLP23 by the failover function (step S16).

  Next, the FW update unit 64 logs in to VLP # 0 (step S17), starts FU of VLP # 0 (step S18), and waits for completion of FU (No in steps S19, S20, and S20). When the FU of VLP # 0 is completed (Yes in step S20), the FW update unit 64 sets VLP # 0 to the master VLP23 by the failover function (step S21).

  Then, the FW update unit 64 sets both the maintenance modes of VLP # 0 and # 1 to OFF (step S22), and the process ends.

[1-4-3] Operation Example of FW Update Process of Back End Unit Next, an operation example of the FW update process of the back end unit 12 shown in Step S5 of FIG. 11 will be described with reference to FIG.

  As illustrated in FIG. 13, the monitoring device 6 sets the maintenance modes of the ICP 25, IDP 26, and tape library 3 to ON (step S31), and logs in to VLP # 0 (step S32).

  Next, the information acquisition unit 62 and the preload control unit 63 of the monitoring device 6 perform preload determination / FU execution determination processing (step S33), and determine whether or not to perform FU (step S34). If it is determined not to perform FU (No in step S34), the process proceeds to step S40. In this case, the operator may adjust the execution timing of the FU and perform the FW update process of FIG.

  If it is determined in step S34 that FU is to be executed (Yes in step S34), the FW update unit 64 stops the PLS process of the VLP 23 and the PDS process of the IDP 26 (step S35).

  Next, the FW update unit 64 performs the FU of the tape library 3, PLP 27, and IDP 26 (steps S36 to S38), and starts the PLS / PDS process after the FU is completed (step S39).

  Then, the FW update unit 64 sets the maintenance mode of each of the IDP 26, the PLP 27, and the tape library 3 to OFF (step S40), and the process ends.

[1-4-4] Operation Example of Preload Determination / FU Implementation Determination Process Next, an operation example of the preload determination / FU execution determination process shown in Step S33 of FIG. 13 will be described with reference to FIG.

  The information acquisition unit 62 of the monitoring device 6 acquires the host access time of each LV 70 of the tape library 3 (step S41). For example, the information acquisition unit 62 acquires the LV list information 611 from the TVC 21. The information acquisition unit 62 may generate the recall history information 612 from the LV list information 611.

  The preload control unit 63 records the acquired last host access time of each LV 70 (step S42), and compares the recorded last host access time with the current time (step S43).

  The preload control unit 63 extracts a candidate LV 70 that is recalled within a predetermined time from the current time (step S44), and selects one candidate LV 70 that is recalled most recently from the extracted candidate LVs 70 (not preloaded) ( Step S45).

  Then, the preload control unit 63 preloads the selected candidate LV 70 to the TVC 21 (step S46).

  The preload control unit 63 determines whether or not all of the extracted candidate LVs 70 have been preloaded (step S47). If not all have been preloaded (No in step S47), the process proceeds to step S45.

  On the other hand, when all the extracted candidate LVs 70 are preloaded (Yes in step S47), the preload control unit 63 determines whether there is an LV 70 that is recalled within a predetermined time from the current time (step S48). If there is an LV 70 that is recalled within a predetermined time from the current time (Yes in step S48), the process proceeds to step S44.

  If there is no LV 70 that is recalled within the predetermined time from the current time in step S48 (No in step S48), the preload control unit 63 calculates the time from the current time until the next LV 70 recall occurs (step S49). ).

  Then, the preload control unit 63 notifies the operator of the calculated time as the FU executable time (step S50), and the process ends.

[1-4-5] Operation Example of Front End Unit FW Update Process Next, an example of FW update process operation of the front end unit 11 will be described with reference to FIGS. 15 and 16. The FW update process of the front end unit 11 is performed, for example, after the process shown in FIG. 11 is completed, continuously or at a later time (for example, at a later date) according to the judgment of the monitoring device 6 or the operator based on the FW executable time. It's okay.

  As illustrated in FIG. 15, the operator adjusts the schedule of the FU work with the user of the virtual tape system 1 (step S51). The schedule includes, for example, the date and time of the FU work of the front end unit 11 and the work period, in other words, the stop date and time and the stop period of the virtual tape system 1.

  Next, the operator confirms the stop of the virtual tape device 2 via the monitoring device 6 (step S52). The stop confirmation may include, for example, confirmation of the presence / absence of I / O from the host device 4 in the virtual tape device 2. If there is an I / O (Yes in steps S53 and S53), the virtual tape device 2 is not stopped, so the process proceeds to step S52, and steps S52 and S53 are executed until there is no I / O.

  If there is no I / O (No in step S53), the virtual tape device 2 is stopped, so the operator performs FW update processing for each component of the front end unit 11 of the virtual tape system 1 (steps S54 and S53). S55). When the FW update process for all the components of the front end unit 11 is completed, the FU operation ends.

  Note that the processing in steps S54 and S55 assumes a case where two operators work in parallel, as illustrated in FIGS. 15 and 16, respectively. The example of FIG. 15 shows the case where the redundant components are FU one by one by the operators A and B. In the example of FIG. 16, the FU work of different types of components is performed in parallel by the operators A and B. The case where it is done is shown.

  The operation example of the FW update process for each component may be the same as the operation example according to the comparative example illustrated in FIG.

  As described above, according to one embodiment, for example, the monitoring device 6 can obtain information on the LV 70 from the VLP 23 using the state investigation tool, and thereby predict a component that does not operate within a certain time during business. it can. As a result, it is possible to preferentially FU a component that is predicted not to operate within a certain time during business. Accordingly, since the components other than the inactive FU can be FU in advance (for example, during business), the business stop time can be reduced.

[1-5] Hardware Configuration Example Next, a hardware configuration example of the monitoring device 6 and the VLP 23 in the virtual tape system 1 shown in FIG. 1 will be described with reference to FIG. Note that both the monitoring device 6 and the VLP 23 can have the same hardware configuration. Hereinafter, for convenience, the monitoring device 6 and the VLP 23 are collectively referred to as a computer 10, and a hardware configuration example of the computer 10 will be described.

  As illustrated in FIG. 17, the computer 10 as an example of the information processing apparatus may include a processor 10a, a memory 10b, a storage unit 10c, an IF unit 10d, an I / O unit 10e, and a reading unit 10f. .

  The processor 10a is an example of an arithmetic processing device that performs various controls and arithmetic operations. The processor 10a may be communicably connected to each of the blocks 10b to 10f via a bus 10i. An integrated circuit (IC) such as a CPU, GPU, MPU, DSP, ASIC, or PLD (for example, FPGA) may be used as the processor 10a. CPU is an abbreviation for Central Processing Unit, GPU is an abbreviation for Graphics Processing Unit, and MPU is an abbreviation for Micro Processing Unit. DSP is an abbreviation for Digital Signal Processor, and ASIC is an abbreviation for Application Specific Integrated Circuit. PLD is an abbreviation for Programmable Logic Device, and FPGA is an abbreviation for Field Programmable Gate Array.

  The memory 10b is an example of hardware that stores various data and programs. Examples of the memory 10b include volatile memory, for example, RAM such as DRAM (Dynamic RAM). Note that RAM is an abbreviation for Random Access Memory.

  The storage unit 10c is an example of hardware that stores various data, programs, and the like. For example, the storage unit 10c may be used as a secondary storage device of the computer 10, and may store an OS, firmware, a program such as an application, and various data. Examples of the storage unit 10c include a magnetic disk device such as an HDD, a semiconductor drive device such as an SSD, and various storage devices such as a nonvolatile memory. Examples of the non-volatile memory include flash memory, SCM (Storage Class Memory), ROM (Read Only Memory), and the like. The storage unit 10c may store a program 10g that realizes all or some of the various functions of the computer 10.

  The IF unit 10d is an example of a communication interface that performs connection control and communication control with other devices via a network. For example, as the IF unit 10d, an adapter conforming to Ethernet (registered trademark), optical communication (for example, FC) or the like can be cited. The computer 10 may have a communication interface for controlling connection and communication with the management terminal of the administrator, etc., and downloading the program 10g from a network (not shown) using the communication interface. Good.

  The I / O unit 10e may include at least one of input devices such as a mouse, a keyboard, a touch panel, and operation buttons, and an output device such as a display, a projector, and a printer.

  The reading unit 10f is an example of a reader that reads data and programs recorded in the recording medium 10h and outputs the data and programs to the processor 10a. The reading unit 10f may include a connection terminal or a device that can connect or insert the recording medium 10h. Examples of the reading unit 10f include an adapter based on USB (Universal Serial Bus), a drive device that accesses a recording disk, a card reader that accesses a flash memory such as an SD card, and the like. Note that the program 10g and the like may be stored in the recording medium 10h.

  Examples of the recording medium 10h include non-transitory computer-readable recording media such as a magnetic / optical disk and a flash memory. Examples of the magnetic / optical disc include a flexible disc, a CD (Compact Disc), a DVD (Digital Versatile Disc), a Blu-ray disc, and an HVD (Holographic Versatile Disc). Examples of the flash memory include a semiconductor memory such as a USB memory and an SD card. Examples of the CD include CD-ROM, CD-R, CD-RW, and the like. Examples of the DVD include a DVD-ROM, a DVD-RAM, a DVD-R, a DVD-RW, a DVD + R, and a DVD + RW.

  The hardware configuration of the computer 10 described above is an example. Therefore, hardware increase / decrease (for example, addition or deletion of an arbitrary block), division, integration in an arbitrary combination, addition or omission of a bus in the computer 10 may be appropriately performed.

  The hardware configurations of the monitoring device 6 and the VLP 23 described above are examples. Therefore, hardware increase / decrease in the monitoring device 6 or the VLP 23 (for example, addition or omission of arbitrary blocks), division, integration in an arbitrary combination, addition or omission of buses, etc. may be performed as appropriate.

  Also, components such as ICP 25, IDP 26, and PLP 27, and terminals such as operation terminal 5 may have the same hardware configuration as that of computer 10 shown in FIG.

[2] Others The technology according to the above-described embodiment can be implemented with modifications and changes as follows.

  For example, the functional blocks of the monitoring device 6 or the VLP 23 shown in FIG. 8 may be merged in an arbitrary combination or may be divided.

  In the embodiment, the monitoring device 6 performs information acquisition, preload control, and FW update. However, the present invention is not limited to this. At least some of the functions of the monitoring device 6 may be provided in the operation terminal 5, the VLP 23, or other components. Devices such as the monitoring device 6 and the VLP 23 that execute at least part of the functions of the monitoring device 6 may be positioned as “control devices” of the virtual tape system 1.

  In one embodiment, the virtual tape system 1 has been described as an example of a storage device, but the present invention is not limited to this. For example, the technique according to the embodiment can be applied to a disk storage system or the like that includes a cache memory as a cache device and a storage such as an HDD or an SSD as a storage device.

[3] Supplementary Notes Regarding the above embodiment, the following supplementary notes are further disclosed.

(Appendix 1)
A front end portion having a cache device;
A back-end unit having a storage device;
A prediction unit that predicts data requested by the cache device within a predetermined time based on a history of access to the cache device;
A storage unit that performs control of reading predicted data from the storage device and storing the data in the cache device;
A storage apparatus, comprising: an update unit that controls updating of software applied to the back end unit in a state where the front end unit is operating.

(Appendix 2)
The prediction unit compares the current time with the access time included in the access history, and predicts the data that the next access to the cache device will occur within the predetermined time.
The storage device according to attachment 1, wherein

(Appendix 3)
The data requested by the cache device is data to be accessed at a predetermined cycle.
The storage apparatus according to Supplementary Note 1 or Supplementary Note 2, wherein:

(Appendix 4)
After the prediction unit stores the predicted data in the cache device, based on the access history, a request for storing data from the storage device to the cache device is generated from the current time to the next time. Predict the updatable time, which is the time,
Whether or not the software applied to the back end unit can be updated is determined based on the updatable time.
The storage apparatus according to any one of appendices 1 to 3, wherein:

(Appendix 5)
The predetermined time is a time equal to or longer than a software update time applied to the backend unit.
The storage device according to any one of appendices 1 to 4, characterized in that:

(Appendix 6)
When a request for storing data from the storage device to the cache device occurs during the update of software applied to the back-end unit, the device included in the front-end unit sends the request to the issuer of the request. Responds that the device performing the processing related to is in use,
The storage device according to any one of appendices 1 to 5, characterized in that:

(Appendix 7)
In a storage device comprising a front-end unit having a cache device and a back-end unit having a storage device,
On the computer,
Predicting data required by the cache device within a predetermined time based on a history of accesses to the cache device;
Control to read the predicted data from the storage device and store it in the cache device;
In a state where the front end unit is in operation, control the update of software applied to the back end unit,
A control program for executing a process.

(Appendix 8)
The prediction compares a current time with an access time included in the access history, and predicts data that will cause a next access to the cache device within the predetermined time.
The control program according to appendix 7, wherein

(Appendix 9)
The data requested by the cache device is data to be accessed at a predetermined cycle.
The control program according to appendix 7 or appendix 8, characterized in that.

(Appendix 10)
The prediction is the time from when the predicted data is stored in the cache device to the next request for storing data from the storage device to the cache device based on the access history. Predict the updatable time,
Whether or not the software applied to the back end unit can be updated is determined based on the updatable time.
The control program according to any one of appendices 7 to 9, characterized in that:

(Appendix 11)
The predetermined time is a time equal to or longer than a software update time applied to the backend unit.
The control program according to any one of appendices 7 to 10, characterized by:

(Appendix 12)
When a request for storing data from the storage device to the cache device occurs during the update of software applied to the back-end unit, the request is issued to the request issuer by the device of the front-end unit. It is replied that the device that performs the processing related to is in use,
The control program according to any one of appendices 7 to 11, characterized in that:

(Appendix 13)
In a storage device comprising a front-end unit having a cache device and a back-end unit having a storage device,
Computer
Predicting data required by the cache device within a predetermined time based on a history of accesses to the cache device;
Control to read the predicted data from the storage device and store it in the cache device;
In a state where the front end unit is in operation, control the update of software applied to the back end unit,
The control method characterized by the above-mentioned.

(Appendix 14)
The prediction compares a current time with an access time included in the access history, and predicts data that will cause a next access to the cache device within the predetermined time.
14. The control method according to appendix 13, wherein

(Appendix 15)
The data requested by the cache device is data to be accessed at a predetermined cycle.
The control method according to Supplementary Note 13 or Supplementary Note 14, wherein

(Appendix 16)
The prediction is the time from when the predicted data is stored in the cache device to the next request for storing data from the storage device to the cache device based on the access history. Predict the updatable time,
Whether or not the software applied to the back end unit can be updated is determined based on the updatable time.
The control method according to any one of appendices 13 to 15, characterized in that:

(Appendix 17)
The predetermined time is a time equal to or longer than a software update time applied to the backend unit.
The control method according to any one of supplementary notes 13 to 16, characterized in that:

(Appendix 18)
When a request for storing data from the storage device to the cache device occurs during the update of software applied to the back-end unit, the device included in the front-end unit sends the request to the issuer of the request. Responds that the device performing the processing related to is in use,
The control method according to any one of appendices 13 to 17, characterized in that:

1 Virtual Tape System 2 Virtual Tape Device 21 TVC
22 FC switch 23 VLP
231 LV information transmission unit 232 Preload execution unit 24 LAN hub 25 ICP
26 IDP
27 PLP
28 PCU
3 Tape Library 31 Drive 32 Robot 4 Host Device 41 VTCP
5 Operation terminal 6 Monitoring device 61 Memory unit 611 LV list information 612 Recall history information 62 Information acquisition unit 63 Preload control unit 64 FW update unit 70 LV
80 PV

Claims (7)

A front end portion having a cache device;
A back-end unit having a storage device;
A prediction unit that predicts data requested by the cache device within a predetermined time based on a history of access to the cache device;
A storage unit that performs control of reading predicted data from the storage device and storing the data in the cache device;
A storage apparatus, comprising: an update unit that controls updating of software applied to the back end unit in a state where the front end unit is operating. The prediction unit compares the current time with the access time included in the access history, and predicts the data that the next access to the cache device will occur within the predetermined time.
The storage apparatus according to claim 1, wherein: The data requested by the cache device is data to be accessed at a predetermined cycle.
The storage apparatus according to claim 1, wherein the storage apparatus is a storage apparatus. After the prediction unit stores the predicted data in the cache device, based on the access history, a request for storing data from the storage device to the cache device is generated from the current time to the next time. Predict the updatable time, which is the time,
Whether or not the software applied to the back end unit can be updated is determined based on the updatable time.
The storage apparatus according to any one of claims 1 to 3, wherein the storage apparatus is characterized in that When a request for storing data from the storage device to the cache device occurs during the update of software applied to the back-end unit, the device included in the front-end unit sends the request to the issuer of the request. Responds that the device performing the processing related to is in use,
The storage apparatus according to any one of claims 1 to 4, wherein the storage apparatus is characterized in that In a storage device comprising a front-end unit having a cache device and a back-end unit having a storage device,
On the computer,
Predicting data required by the cache device within a predetermined time based on a history of accesses to the cache device;
Control to read the predicted data from the storage device and store it in the cache device;
In a state where the front end unit is in operation, control the update of software applied to the back end unit,
A control program for executing a process. In a storage device comprising a front-end unit having a cache device and a back-end unit having a storage device,
Computer
Predicting data required by the cache device within a predetermined time based on a history of accesses to the cache device;
Control to read the predicted data from the storage device and store it in the cache device;
In a state where the front end unit is in operation, control the update of software applied to the back end unit,
The control method characterized by the above-mentioned.

Images