JPWO2016001959A1 - Storage system - Google Patents

Abstract

Since the prefetch operation takes time, if the prefetch timing is not scheduled, copying (prefetch) from a low-speed storage medium to a high-speed storage medium cannot be executed when necessary, resulting in a processing wait. . In one embodiment of the present invention, a prefetch schedule for copying data from a first storage medium to a second storage medium in advance before a read request from an external computer is determined based on the time taken for the prefetch.

Description

  The present invention relates to drive control in a storage system.

  In the storage system, to improve the Read performance, before executing the application (AP), copy the data from the storage medium storing the data to a storage device that can be read at higher speed (this is called prefetch) to create cache data To do. For example, in a storage system disclosed in Patent Document 1 including a magnetic disk and an optical disk, there is a technique for creating cache data by copying data on the optical disk to the magnetic disk according to a request tendency.

US Pat. No. 6,233,653

  The prefetch operation also takes time, so if the prefetch timing is not scheduled, copying (prefetch) from a low-speed storage medium to a high-speed storage medium cannot be executed when necessary, and processing waits will occur. There is.

  In order to solve the above problem, in one embodiment of the present invention, a prefetch schedule for copying data from a first storage medium to a second storage medium in advance before a read request from an external computer is applied to the prefetch. Decide based on time.

  Response time to the host device is shortened.

FIG. 1 is a diagram for explaining the outline of the embodiment. FIG. 2 is an overall configuration diagram of the first embodiment. FIG. 3A is a configuration diagram of the host computer according to the first embodiment. FIG. 3B is a configuration diagram of the file server according to the first embodiment. FIG. 4 is a configuration diagram of the block storage according to the first embodiment. FIG. 5 is a configuration diagram of the portable storage device according to the first embodiment. FIG. 6A is a configuration diagram of the hybrid control server according to the first embodiment. FIG. 6B is a configuration diagram of the job management server according to the first embodiment. FIG. 7 is a configuration diagram of the apparatus management server according to the first embodiment. FIG. 8 is a configuration diagram of the archive management server according to the first embodiment. FIG. 9A is a diagram illustrating information in the file configuration management table of the first embodiment. FIG. 9B is a diagram illustrating information of the block configuration management table according to the first embodiment. FIG. 9C is a diagram illustrating information of the portable storage device configuration management table according to the first embodiment. FIG. 10 is a diagram illustrating information of the job management table according to the first embodiment. FIG. 11 is a diagram illustrating information of the configuration management table according to the first embodiment. FIG. 12 is a diagram illustrating information in the data management table of the first embodiment. FIG. 13 is a diagram illustrating information in the job correspondence management table according to the first embodiment. FIG. 14A is a diagram illustrating information of the HDD capacity management table according to the first embodiment. FIG. 14B is a diagram illustrating information of the policy management table according to the first embodiment. FIG. 15 is a diagram illustrating an overall flow of the first embodiment. FIG. 16 is a flowchart illustrating a job time calculation flow of the drive use process according to the first embodiment. FIG. 17 is a diagram illustrating a flow of determining the job execution order and the used drive according to the first embodiment. FIG. 18 is a diagram showing a flow of merge processing when there is another job using the same medium. FIG. 19 is a diagram illustrating a prefetch data deletion flow according to the first embodiment. FIG. 20 is a flowchart illustrating the job time calculation flow of the drive use process according to the second embodiment. FIG. 21 is a diagram illustrating a flow of determining the job execution order and the used drive according to the second embodiment. FIG. 22 is a diagram illustrating a flow of information collection processing for HDD capacity according to the third embodiment. FIG. 23 is a diagram illustrating a flow of determining the job execution order and the used drive according to the fourth embodiment. FIG. 24A is a diagram illustrating information of the portable storage device configuration management table according to the fifth embodiment. FIG. 24B is a diagram illustrating information in the offline portable storage medium management table according to the fifth embodiment. FIG. 25 is a flowchart illustrating a job time calculation flow of the drive use processing according to the fifth embodiment.

  FIG. 1 is a diagram showing an outline of this embodiment. The device management server 65 holds a storage management program 504 and a drive use processing calculation program 511. The portable storage device 40 includes a drive 405 that records and reproduces data, a medium transport unit 406, and a plurality of portable storage media 408. The hybrid storage device 91 includes a hybrid control server 50, a block storage 31, and a portable storage device 40. The block storage 31 has a logical volume 313. The hybrid control server 50 has a read / write program 512.

  The hybrid storage device 91 uses the block storage 31 and the portable storage device 40 to archive data. Data from the higher-level storage device is first written to the block storage 31 by the read / write program 512 of the hybrid control server 50, and written to the portable storage device 40 by the read / write program 512 when a specific condition is satisfied. When reading data, the read / write program 512 reads data from the portable storage device 40 and copies it to the block storage 31. If this copy process is performed (prefetched) before the time required by the application, it becomes possible to conceal the read delay due to the medium transport time of the portable storage device 40.

  The storage management program 504 collects various information from the host computer 10, the file server 20, the block storage 30, the portable storage device 40, the archive management server 70, and the job management server 60 periodically or at the timing of data generation. Next, a schedule for loading the plurality of portable storage media into the drive is determined based on application schedule information, block storage storage information, drive performance information, and data storage location information included in the collected data.

  According to the present embodiment, it is possible to determine an appropriate execution order of processing using the drive in the portable storage device 40. As a result, it is possible to prevent delays in Read / Write processing using the portable storage device 40. This is particularly effective when the number of drives is small or when the portable storage medium 408 has a low Read / Write performance such as a hologram disk. The effect of this embodiment is important when the portable storage device 40 is used for the purpose of Active Archive in which both Read / Write processes occur at the same frequency.

  FIG. 2 is a block diagram showing a configuration example of the entire computer system 1. The host computer 10 is connected to the file server 20 and the hybrid storage device 91 via the data network 80a. The hybrid storage device 91 includes a hybrid control server 50, a block storage 31, and a portable storage device 40. The hybrid control server 50 and the block storage 31 are connected via a data network 80c. The hybrid control server 50 and the portable storage device 40 are connected via a data network 80d. The file server 20 and the block storage 30 are connected via a data network 80b. The host computer 10, file server 20, block storage 30, hybrid storage device 91 (hybrid control server 50, block storage 31, portable storage device 40), job management server 60, device management server 65, and archive management server 70 are managed. The network 90 is connected. The data networks 80a, 80b, 80c, and 80d do not need to be separated, and may be the same network. The protocol of the data network 80 and the management network 90 may be any protocol such as FC (Fibre Channel) or IP (Internet Protocol), and the data network 80 and the management network 90 may be the same network. A system including the hybrid storage device 91 and the device management server 65 is collectively referred to as a storage system 92. Further, a system including the storage system 92, the file server 20, the block storage 30, the job management server 60, and the archive management server 70 is collectively referred to as a hierarchical storage system 93.

  In the present embodiment, a case where each server is independent is shown, but each server is not necessarily independent, and each process may be executed by one or more servers. Further, a storage device or a portable storage device may execute the processing of each server. In addition, although an example in which the storage device 30 and the storage device 31 are different cases has been shown, the same case may be used, and a storage area may be provided for each of the file server 20 and the hybrid control server 50.

  FIG. 3A is a diagram illustrating an example of the host computer 10. The host computer 10 includes a CPU 101, a memory 102, a data IF (Interface) 108, and a management IF 109. The memory 102 stores an OS (Operating System) 103 and a device manager 107 for mounting a storage area of the storage apparatus. The OS 103 includes an application 104, a hierarchy management agent 105, and a job execution agent 106. The CPU 101 operates the OS 103 and the device manager 107 on the memory. The hierarchy management agent 105 communicates with a hierarchy management program 704 in the archive management server 70 described later. The job execution agent 106 communicates with a job management program 604 of the job management server 60 described later. The data IF 108 is connected to the data network 80. The management interface 109 is connected to the management network 90. The data interface 108 and the management interface 109 may be a common IF.

  FIG. 3B is a diagram illustrating an example of the file server 20. The file server 20 includes a host IF 201, a file control processor 202, a memory 203, a disk IF 206, and a management IF 207. The memory 203 stores a file configuration management table 204 and a hierarchy management agent 205. The hierarchy management agent 205 communicates with the hierarchy management program 704 in the archive management server 70. The file control processor 202 operates the hierarchical management agent in the memory. The host IF 201 is connected to the host computer 10 via the data network 80a. The disk IF 206 is connected to the block storage 30 via the data network 80b. The management interface 207 is connected to the device management server 65 via the management network 90. The file control processor 202 mounts the volume of the block storage 30 and causes the file server 20 to operate as NAS (Network Attached Storage).

  FIG. 4 is a diagram illustrating an example of the block storages 30 and 31. The block storages 30 and 31 include a block control processor 302, a memory 303, a physical storage device 306, and a management IF 315. The memory 303 stores a block configuration table 304 and a hierarchy management agent 305. The hierarchy management agent 305 communicates with the hierarchy management program 704 in the archive management server 70. The block control processor 302 operates the hierarchical management agent 305 in the memory. Details of the block configuration table 304 will be described later. The physical storage device 306 is a storage medium such as one or more HDDs (Hard Disk Drives) 307 and one or more SSDs (Solid State Drives) 308. A plurality of types of physical storage areas may be included. Note that the type of physical storage device is not limited to HDD or SSD, but may be any other type of device.

  Here, the physical storage device of the physical block storage 31 may be referred to as a prefetch storage medium because data may be stored for the purpose of prefetching from the storage medium 408 of the portable storage device. Since this prefetch storage medium is used for prefetching, it is expected that the access performance is higher than that of the storage medium 408. Storage media A plurality of physical storage devices have a RAID (Redundant Arrays of Independent Disks) configuration. The physical control processor 302 provides the logical volume 312 to the host computer 10 and the file server 20 based on the physical storage device. The management interface 315 is connected to the device management server 65 via the management network 90.

  FIG. 5 is a diagram illustrating an example of a portable storage device. The portable storage device is a storage device that sets a portable storage medium in a drive using a medium transport mechanism such as a robot. The portable storage device 40 includes a host IF 401, a controller 402, a memory 403, a cache memory 404, a drive 405, a medium transport unit 406, and a medium storage unit 407. The host IF 401 is connected to the host computer 10 via the data network 80. The management IF 409 is connected to the device management server 65 via the management network 90.

  The memory 403 holds a Read / Write program 410, an agent program 411, a portable storage device management program 412, and a portable storage device configuration management table 413. The Read / Write program 410 is a program for writing data to the storage medium 408 through the drive 405 or reproducing data on the medium through the drive 405. The agent program 411 is a program for communicating with the storage management program 654 and the read / write instruction program 657 of the device management server 65. The portable storage device management program 412 is a program that manages the medium transport unit 406 that transports the storage medium 408 included in the medium storage unit 407 and manages information on the portable storage medium 408. Details of the portable storage device configuration management table 413 will be described later. The controller 402 operates the program in the memory 403. The cache memory 404 is a storage medium that temporarily caches data sent from an upper layer and data read from the portable storage medium 408, and is composed of a DRAM, a flash HDD, etc., and has a read / write performance that is portable storage. Higher than medium.

  The drive 405 reads / writes data from / to the storage medium 408 transported by the medium transport unit 406. The number of drives 405 is one or more and less than the number of storage media 408. Further, when a plurality of drives 405 are provided, drives having different performances may be mixed. Further, when there are a plurality of types of storage media 408, there may be a dedicated drive 405 for each type of storage medium, or there may be a common drive 405 for different types of storage media.

  The medium transport unit 406 transports the storage medium 408 with the storage medium storage unit 407 and the drive 405. Here, arranging the storage medium 408 in a readable state in the drive 405 is called loading. The number of storage medium transport units 406 may be one or more, and is not necessarily equal to the number of drives. In this embodiment, all medium transport units 407 can transport storage media to all drives 405 and medium storage units 407. However, a specific medium transport unit 407 can be connected between a specific medium drive and a medium storage unit. Only conveyance may be performed. The medium storage unit 407 stores a plurality of storage media 408. There may be one or more medium storage units 407. The portable storage medium 408 is an optical storage medium such as one or more DVDs (Digital Versatile Discs), Blu-ray (registered trademark) Discs and hologram disks, and a magnetic storage medium such as tapes. The number of storage media 408 is greater than the number of drives 405, and a plurality of types of storage media may be mixed.

  FIG. 6A is a block diagram illustrating an example of the hybrid control server 50. The hybrid control server 50 includes a host IF, a file control processor 502, a memory 503, a disk IF 504, a library IF 505, and a management IF 506.

  The hybrid control server 50 provides the block storage 31 and the portable storage device 40 as a single storage device to the host computer. Specifically, the hybrid control server 50 is provided to the host computer as a storage device having the storage capacity of the portable storage device 40, and the storage area of the block storage 31 is used as a cache area accessed by the host.

  The memory 503 includes a file configuration management table 501, a hierarchy management agent 511, and a Read / Write program 512. The hierarchy management agent 511 communicates with the hierarchy management program 704 in the archive management server 70. The file control processor 502 operates the hierarchical management agent in the memory. The host IF 501 is connected to the host computer 10 via the data network 80a. The disk IF 504 is connected to the block storage 30 via the data network 80c. The library IF 505 is connected to the portable storage device 40 via the data network 80d. The management IF 506 is connected to the device management server 65 and the archive management server 70 via the management network 90.

  The file control processor 502 mounts the logical volume of the block storage 31 and the storage medium or volume of the portable storage device 40, and operates the hybrid control server 50 as NAS (Network Attached Storage). Here, detailed operation of NAS is omitted. The file configuration management table 510 in the memory 503 has physical resource information that the hybrid control server 50 has. Details of the file configuration management table are the same as those of the file configuration management table 204 of the file management server 20 described later. The Read / Write program 512 is a program that reads / writes data in the block storage 31 or the portable storage device 40 according to an instruction from the host computer 10 or the hierarchy management program 70.

  FIG. 6B is a block diagram illustrating an example of the job management server 60. The job management server 60 includes a CPU 601, a NIC 602, and a memory 603. The memory 603 stores a job management program 604 and a job management table 605. A job management program 604 is a program that executes job processing that operates in the hierarchical management system. For job execution, the planned job is executed by communicating with the job execution agent 106 of the host computer 10, the job execution agent 663 of the apparatus management server 65, and the job execution agent 707 of the archive management server 70. Details of the job management table 605 will be described later. Further, it is connected to the management network 90 via the NIC 602. An input / output device 620 for operating the job management server 60 is connected to the job management server 60. The input / output device 620 is a device such as a mouse, a keyboard, and a display, and is used to input / output information between the job management server 60 and an administrator (also referred to as a user).

  FIG. 7 is a block diagram illustrating an example of the device management server 65. The management network 90 is connected to the host computer 10, the file server 20, the block storage 30, the hybrid storage device 91 (hybrid control server 50, block storage 31, portable storage device 40), job management server 60, and archive management server 70. Thus, necessary information can be obtained from each computer, and necessary information (including a program) can be given to each computer.

  The device management server 65 includes a CPU 651, a NIC 652, and a memory 653, to which an input / output device 670 is connected. The CPU 651 executes a program developed on the memory 653. The memory 653 includes a storage management program 654, a system management program 655, a drive use processing calculation program 656, a read / write instruction program 657, a configuration management table 658, a data management table 659, a job correspondence management table 660, an HDD capacity management table 661, a policy. A management table 662, a job execution agent 663, and an execution script 664 are stored. Other programs may be stored.

  The storage management program 654 uses the information in the configuration management table 658, the data management table 659, the job correspondence management table 660, the HDD capacity management table 661, and the policy management table 662 to manage each management target (host computer 10, file server). 20, the block storage 30, the hybrid storage device 91 (hybrid control server 50, block storage 31, portable storage device 40), job management server 60, and archive management server 70) are managed. In addition, the storage management program 654 can communicate with the agent program 411 in the portable storage device 40 to receive instructions and notification of information. The system management program 655 can communicate with the job management program 604 of the job management server 60 and the hierarchy management program 704 of the archive management server 70 to receive instructions and notification of information. Further, the storage management program 654 and the system management program 655 may be the same program.

  The drive use processing calculation program 656 uses the information in the configuration management table 658, the data management table 659, the job correspondence management table 660, the HDD capacity management table 661, and the policy management table 662 to describe the content of the process that occupies the drive ( Processing time). The read / write instruction program 657 is a program that gives an instruction to write data stored in the cache memory 404 in the portable storage device 40 to the storage medium 408. Details of the configuration management table 658, the data management table 659, the job correspondence management table 660, and the job correspondence management table HDD capacity management table 661 will be described later.

  The policy management table 662 is a table describing a policy for moving data in units of files, host computers, and the like. The same contents as the migration policy table 705 in the archive management server 70 are included. The job execution agent 663 is an agent that receives a command from the job management program 604 and executes processing. The execution script 664 is a script that describes the contents of processing executed by the job execution agent 663. A process for calling a plurality of programs such as a storage management program and a system management program according to the process is described. The input / output device 670 is a device such as a mouse, a keyboard, and a display, and is used to input / output information between the device management server 65 and an administrator (also referred to as a user).

  FIG. 8 is a block diagram illustrating an example of the archive management server 70. Host computer 10, file server 20, block storage 30, hybrid storage device 91 (hybrid control server 50, block storage 31, portable storage device 40), job via archive management server 70 and data network 80 or management network 90 Necessary information can be acquired from each computer via the management network 80 connected to the management server 60 and the device management server 65, and necessary information (including programs) can be given to each computer.

  The archive management server 70 includes a CPU 701, a NIC 702, and a memory 703, which are processors, to which an input / output device 720 is connected. The CPU 701 executes a program developed on the memory 703. The memory 703 stores a hierarchy management program 704, a migration policy table 705, a data management table 706, and a job execution agent 707. Other programs may be stored.

  The hierarchy management program 704 communicates with the hierarchy management agents 105, 205, 305, and 511 to acquire file information of the host computer 10, the file server 20, and the hybrid control server 50, and the host computer 10, the file server 20, and block storage. 30, the block storage 31 connected to the hybrid control server 50 and the file or volume of the portable storage device 40 are transferred to another host computer 10, the file server 20, the block storage 30, and the block storage 31 connected to the hybrid control server 50 and the portable storage device. It is a program that moves to 40.

  The policy management table 705 is a table describing a policy for moving data in units of files used in files, host computers, and the like. The policy management table 705 shows the original contents including the same contents as the movement policy table 662 in the device management server 65. The data management table 706 is a table showing attribute information of files in the host computer 10, the file server 20, and the hybrid control server 50.

  The input / output device 720 is a device such as a mouse, a keyboard, and a display, and is used to input / output information between the device management server 70 and an administrator (also referred to as a user).

FIG. 9A is a file configuration management table 204 in the file server 20 and in the hybrid control server 50. The file configuration management table 204 has the following information (2040) to (2042). Note that the file configuration management table 510 has the same configuration as the file configuration management table 204.
(2040) Device type. Contains information on the device type of the file server.
(2041) Device identifier. It has information that uniquely identifies the device that the file server has.
(2042) Spec. For file systems, the size is set.

FIG. 9B is a block configuration management table 304 in the block storages 30 and 31. The block configuration management table 304 has the following information (3040) to (3042).
(3040) Device type. Contains device type information of block storage. The device manages resources such as ports of the block storage including the physical storage device and the logical volume 312 provided to the host.
(3041) Device identifier. It has information that uniquely identifies the device that the block storage has.
(3042) Spec. In the case of logical volumes having different values set depending on the device type, the size and the read / write speed, the speed in the case of a port, and the read / write speed in the case of a physical storage device are set.

FIG. 9C is a portable storage device configuration management table 41 of the portable storage device 40. The portable storage device configuration management table 413 has the following information (4130) to (4134).
(4130) Device type. It has information on the type of device that the portable storage device has. Here, the device includes not only the storage medium 408 but also a cache, a drive, and a medium transport unit.
(4131) Device identifier. It has information that uniquely identifies the device that the portable storage device has.
(4132) Specification. In the case of a cache, the size is set. In the case of a storage medium, the size is set.
(4133) Write unit. Indicates the minimum size that can be written to a storage medium such as an optical disc at one time.
(4134) Media expiration date. In the case of a hologram disc, there is a characteristic that writing becomes impossible unless writing is completed within a certain period. This is referred to as the medium expiration date or shelf life. Indicates the date of the deadline. The remaining time may be held instead of the date.

FIG. 10 shows a job management table 605. The job management table has the following information (6050) to (6057). Information in the job management table is set by user input from the input / output device 620 of the job management server 60 or job calculation by a drive use processing calculation program.
(6050) A job identifier. Has information that uniquely identifies the job.
(6051) Job type. It has information indicating whether the content of the job is Read / Write.
(6052) A job execution agent machine. Contains information about the machine name of the machine that executes each job. Here, Host represents a host computer, and Mgmt represents a management server.
(6053) Job execution agent IP address. It has information on the IP address of the machine that executes each job.
(6054) Application / script used. Contains information about the location of the app or script on the machine that is executed by each job.
(6055) Usage data. It has data information used by the application or script on the machine executed by each job. Data information is a file or directory name recognized by the host computer or application.
(6056) Scheduled job execution start time. Contains information about the time when the job is scheduled to be executed.
(6057) Job execution completion required time. Contains information about the time at which the job should be completed. If the user does not set, it may be blank.

  FIG. 11 shows a configuration management table 658 provided in the device management server 65. The configuration management table has information (6580) to (6587). Information of the configuration management table is generated from the file configuration management table 204 in the file server (the file configuration management table 510 in the hybrid control server 50), the block configuration management table 304 in the block storage, and the portable storage device configuration management table 413. Is done. These pieces of information are periodically collected from the management target storing each piece of information by the storage management program 654. Details of each column of the configuration management table 658 are omitted here.

FIG. 12 is an explanation of the data management table 659 included in the device management server 65. The data management table has the following information (6590) to (6595). The data management table 706 has the same structure and the same information as the data management table 659.
(6590) Data identifier. Indicates information that uniquely identifies data to be managed.
(6591) Data size. Indicates the size of the data.
(6592) Apparatus. Indicates the location of the storage device where the data is stored.
(6593) In-device storage location. Indicates the location of a logical storage area in the storage device where the data is stored.
(6594) Data path. A data path in a storage area in a storage device in which data is stored is shown.
(6595) Last access time. Indicates the time when the data was last accessed.

FIG. 13 is a job correspondence right management table 660 provided in the apparatus management server 65. The job correspondence management table is a table for managing the correspondence between the job of the upper application executed by the host computer or the management computer and the drive use job that is the job executed by the drive of the portable storage device in order to execute the upper application. It is. The job correspondence management table job has the following information (6600) to (6610).
(6600) Job of upper application. The identifier of the job scheduled to be executed by the host application, specified by the host computer or management computer.
(6601) Indicates whether the content of the job is Read / Write.
(6602) Usage data. It has data information used by the application or script on the machine executed by each job.
(6603) Scheduled execution start time of job of upper AP. It has information on the time when the upper AP job is scheduled to be executed.
(6604) Mandatory completion time of the upper AP job. Contains information about the time at which the job should be completed. If the user does not set, it may be blank.
(6605) Drive use job. It has job identifier information input by the user or calculated by the drive use processing calculation program.
(6606) A required portable storage medium. It has information on the identifier of the storage medium used by the drive use job in Read / Write processing.
(6607) Drive scheduled to be used. The drive use job has information on the identifier of the drive to be used.
(6608) Drive occupation start scheduled time. Contains information about the time when a drive use job is scheduled to start execution.
(6609) Scheduled end time of drive occupation. Contains information about the scheduled end time of a job that uses a drive.
(6610) Related drive use job. When related to another drive use job, it has information on the identifier of the job. Here, the “related job” means a job in which post-processing such as data deletion cannot be executed unless the processing is completed after both storage media used in the job are the same. For example, even if the execution of Job A that requires Job α ends, data deletion is not performed until the execution of Job C that uses Job γ ends. In this embodiment, the drive to be used is managed for the job, but it may be managed whether the job is assigned for each drive time.

FIG. 14A is an HDD capacity management table 661 provided in the device management server 65. The HDD capacity management table has the following information (6611) to (6615).
(6611) Apparatus. The identification information of the device having the HDD is shown.
(6612) Volume. Indicates the identifier of the logical volume in the device.
(6613) Time. Indicates current and future time. It should be noted that an arbitrary time interval may be used, and the current past time may also be managed.
(6614) Total capacity. Indicates the total capacity of the logical volume described in Volume.
(6615) Free capacity. Indicates the free capacity of the logical volume for each hour.

FIG. 14B shows a policy management table 662 provided in the device management server 65. The policy management table 662 has the following information (6620) to (6623). The policy management table 705 has the same structure and the same information as the policy management table 662.
(6620) File. Indicates the file identification information. In this embodiment, a policy is set for each file, but a migration policy may be set for each host or directory.
(6621) Hierarchy. Indicates the storage device where the file exists. Here, the case where it is stored in the block storage 30 is ST1, the case where it is stored in the block storage 31 is ST2, and the case where it is stored in the portable storage device 40 is ST3. In the case where storage media having different properties are provided in the block storage 30, the data stored in the block storage 30 may be divided into a plurality of hierarchies like ST0 and ST1. Further, when the portable storage device 40 stores HDDs and optical disks as storage media having different properties, the portable storage device 40 may be divided into a plurality of layers as in ST3 and ST4.
(6622) Migration policy. Indicates the data movement policy. In the case of FIG. 14B, it is shown that if there is no access for one month to File1 included in ST1, data is moved to another layer called ST2.
(6623) Deletion policy after movement. Indicates the processing contents for the original file after the data is moved according to the policy.

FIG. 15 shows the overall processing flow in the first embodiment. The first embodiment shows a flow of a drive usage time scheduling process when a prefetch process is executed when a higher-level Read job is input.
(Step 1000) Each program periodically collects information. Specifically, first, the user inputs job execution information (job type, job execution machine, application / script used, application data, execution time, execution completion time, etc.) from the input / output device 620 of the job management server 60. Next, the job management table 605 is updated according to the input contents.

  Further, the hierarchy management program 704 of the archive management server 70 periodically collects file information from the OS and application of each host. The hierarchy management program 704 updates the data management table 706.

  In addition, the system management program 655 of the device management server 65 periodically performs file configuration management from the file server 20, block storages 30 and 31, hybrid control server 50, portable storage device 40, job management server 60, and archive management server 70. The table 204, the block management table 304, the portable storage device management table 413, the file configuration management table 510, the job management table 605, the policy management table 705, and the data management table 706 are acquired. These information acquisition frequencies may be acquired at different timings for each device or table.

The storage management program 654 updates the configuration management table 658, the data management table 659, the job correspondence management table 660, the HDD capacity management table 661, and the policy management table 662. If there is a difference between the job information acquired from the job management server 60 by the system management program 655 and the information in the job correspondence management table 660, the difference is notified to the drive utilization processing calculation program 656.
(Step 1100) The drive use processing calculation program 656 calculates the time required for the drive use processing (prefetch processing) for reading the data required for the job.
(Step 1200) The drive use processing calculation program 656 determines a job execution order and a drive to be used in the drive use processing from previously registered job information and information on the drive. Further, the drive use process is registered as a job.
(Step 1300) The job management program 604 executes various job processes.
(Step 1400) Data prefetched by the hierarchy management program 704 is deleted according to the policy.

In FIG. 16, the detailed processing of the drive usage processing calculation in Step 1100 will be described. In the first embodiment, a process for reading data from a storage medium is shown as the drive utilization process.
(Step 1101) The drive use processing calculation program 656 detects the job addition of the higher-level application based on the notification from the system management program or 655.
(Step 1102) The drive use processing calculation program 656 specifies data and time required for the upper application / script from the job correspondence management table 660. Note that data that is an argument of the execution application described in the script may be used.
(Step 1103) The drive utilization processing calculation program 656 checks the position of the target data from the contents of the job correspondence management table 660 and the data management table 659. If the position is the portable storage device 40, the process proceeds to Step 1104. If it is not in the portable storage device 40, the process proceeds to Step 1106.
(Step 1104) The drive use processing calculation program 656 acquires the read performance of the drive, the write performance of the HDD, and the load time from the configuration management table 658.
(Step 1105) The drive use processing calculation program 656 calculates the time required for prefetching (copying from the portable storage device to the block storage HDD) using the data size of the data required and various information acquired in the previous flow. . For example, as the simplest calculation method, a method of adding the time required for reading data, the time required for copying, and the load time can be considered. That is, the time T_r [sec] required for data read is calculated as T_r = D_Size / P_R, where the data size is D_Size [GB] and the read performance of the drive is P_R [GB / sec]. The time T_w [sec] for writing data to the HDD can be calculated as T_w = D_Size / P_W, where the data size is D_Size [GB] and the write performance of the HDD is P_W [GB / sec]. If the disk loading time is T_l [sec], the time required for prefetching can be calculated as T_Pre = T_r + T_w + T_l. In addition, when there are a plurality of target data and a plurality of disks storing the data and a plurality of drives, it is possible to calculate a time required for prefetching by performing a plurality of processes in parallel. In addition, when the write to the HDD is possible after the read of a part of the data is completed, the write may be started before the read is completed, and the prefetch time may be calculated so as to exclude the overlapping time. If the loading time of the disc varies depending on the position, the calculation may be performed taking that into consideration. Alternatively, past calculation results and actual prefetch processing actual times may be recorded, and calculation times may be learned and calculated based on the differences. Further, the prefetch time may be calculated by other calculation methods.
(Step 1106) The drive use processing calculation program 656 determines that the prefetch processing is unnecessary for this job.

The details of the process of determining the job execution order and used drive in Step 1200 will be described with reference to FIG.
(Step 1201) The drive use processing calculation program 656 checks job information occupying the drive from the job correspondence management table 660.
(Step 1202) The drive use processing calculation program 656 uses another portable storage medium that is the same as the portable storage medium used in the current drive use processing job, and another prefetch job is before the job start time of the upper application. Check to see if. If yes, go to Step 1210 of FIG. If there is not, go to Step 1203.
(Step 1203) The drive use processing calculation program 656 searches the job correspondence management table 660 for a drive free time before the batch job start time of the higher-level application that is the source of the prefetch job. In Step 1105, if it is checked whether a plurality of disks and a plurality of drives can be prefetched in parallel, a plurality of drives are set as search targets.
(Step 1204) The drive utilization processing calculation program 656 checks whether there is a time when there is a sufficient HDD free space in the free time searched in Step 1203. If there is a time with sufficient free space, the process proceeds to Step 1206. If not, the process proceeds to Step 1205.
(Step 1205) The drive use processing calculation program 656 returns an error to the job management program 604. As a result, it is possible to display in the input / output device 620 that the job that the user tried to register cannot be executed on time.
(Step 1206) The drive utilization processing calculation program 656 selects a period with sufficient time and the drive, and determines the job order to execute the prefetch job at that time.

  There are several methods as a drive selection method and a job order determination method. One is a method called simple pre-packing. This method prepares a job net (group that can have a set of jobs) for each drive, calculates the time required for drive usage processing, and then assigns the job to the drive with the free time closest to the current time. This is how to set up the net. In the case of this method, it is possible to assign jobs to empty drives from the front in time series.

In addition, since the calculation accuracy of the drive use processing time is not necessarily accurate, there is a method of assigning job execution time to a free time with the longest interval between other jobs instead of always assigning from the front in time. In the case of front-packing, when the job processing of the upper AP is executed in a relatively distant future (one month later, one year later, etc.), the interval between the processing executed in advance and the job execution time of the upper AP is long. Become. In that case, since the capacity of the HDD is occupied more than necessary, there is a method in which a period is provided from the job execution time of the higher-level AP, and the front end is settled within that period. This period may be a fixed value, or may be dynamically changed to an appropriate time by learning from the type of upper AP and past results. Further, there is a method of preferentially selecting a drive with good drive performance or a drive performance that satisfies the prefetch time prediction. Further, the job execution time and drive may be determined by other methods.
(Step 1207) The drive utilization processing calculation program 656 reduces the HDD free capacity during the prefetch data holding period from the HDD free capacity table 661.
(Step 1208) The drive utilization processing calculation program 656 creates a script for prefetching data with the designated drive. This script describes a process in which the hierarchy management program 704 copies the optical disk data to the HDD via the hierarchy management agent of the hybrid control server 50.
(Step 1209) A job for executing a script created by the system management program 655 is registered in the job management table 660 and the job management table 605 via the job management program 604. This completes the process.

FIG. 18 shows merge processing when there is another job using the same medium in S1202.
(Step 1210) The drive utilization process calculation program 656 starts a process of merging a plurality of jobs after Step 1211.
(Step 1211) The drive use processing calculation program 656 compares the portable storage medium used in the previous prefetch job with the portable storage medium used in the current prefetch job.
(Step 1212) The drive use processing calculation program 656 checks whether all the portable storage media used in the current prefetch job are included in the portable storage media used in the previous prefetch job. If all are included, go to Step 1213. When not included, the processing after Step 1213 is performed only on the portable storage medium partially included. For the portable storage medium not included, the processing of Step 1203 is performed.
(Step 1213) The drive utilization processing calculation program 656 converts the data required for the upper job corresponding to the current prefetch job into the prefetch data from the end of the prefetch data holding period of the previous prefetch job until the upper job completion time. The HDD free capacity table 661 is checked to see if it exists. If there is free space, the process proceeds to Step 1214. If not, the process proceeds to Step 1203.
(Step 1214) The drive use processing calculation program 656 reduces the HDD free capacity from the HDD free capacity table 661 from the end of the previous prefetch data holding period to the upper job completion time.
(Step 1215) The drive utilization processing calculation program 656 determines that the current prefetch job corresponds to the previous prefetch job, and registers the job identifier of the previous prefetch job in the related prefetch job in the job correspondence management table 660. At that time, the drive occupation start scheduled time and the drive occupation end scheduled time are left blank.

Details of the prefetch data deletion processing in Step 1400 will be described with reference to FIG.
(Step 1401) The drive utilization processing calculation program 656 inquires of the job management program 604 via the system management program 655 whether or not the job execution of the upper AP has been completed. Alternatively, a job completion notification of the upper AP may be received from the job management program 604. In that case, the process proceeds to Step 1403.
(Step 1402) When the upper-level job is completed, the process proceeds to Step 1403. If not completed, the process returns to Step 1401.
(Step 1403) The drive use processing calculation program 656 determines from the job correspondence management table 660 whether the completed drive use job of the upper AP job is a related drive use job of the higher AP. This is because the data scheduled to be used by the higher-order APs thereafter cannot be deleted. Check if there is an upper AP.
(Step 1404) If there is a higher-order AP that uses the same data as a result of confirmation in Step 1403, the process proceeds to Step 1405. If not, the process proceeds to Step 1406.
(Step 1405) S1400 is executed for the upper AP using the same data.
(Step 1406) The system management program 655 instructs the hierarchy management program 704 to delete data according to the policy management table for the prefetched data.
(Step 1407) The storage management program 654 updates the HDD free capacity in the HDD capacity management table 661.

  In this embodiment, the logical volume of the block storage 31 is used as a portion for caching data, but the cache memory 404 in the portable storage device may be used, or the storage area of the lower storage of the hierarchical storage is used. May be.

  With the above processing, it is possible to schedule the drive use processing of the portable storage device and store the archived file in a capacity-efficient manner, so that the waiting for the read processing is less likely to occur, resulting in waiting for the processing of the upper-level AP. It becomes difficult.

  In the second embodiment, job scheduling including not only the Read process but also the Write process is performed. Only differences from the first embodiment will be described below. Differences from the first embodiment are two sequences: calculation of drive usage processing, and job execution order and usage drive calculation.

The job execution time calculation process of the drive use process in the second embodiment will be described with reference to FIG.
(Step 2101) Similar to Step 1101, the drive use processing calculation program 656 detects the job addition of the upper application. This occurs as a result of Step 1032. Alternatively, the completion of data accumulation in the logical volume of the block storage by the tier management program 704 or the start of writing by the policy is detected and notified to the drive utilization processing calculation program 656. Here, completion of data storage in the logical volume means that data writing from the logical volume to the portable storage medium does not always work with the upper AP, and data is stored after storing data for one optical disk after data is stored. There is a case where data is written according to a policy even if one is not stored. This process is not necessarily performed by the tier management program 704, but may be performed by the storage management program 654 or the system management program 655.
(Step 2102) If the detected content of the job process is a write process, the process advances to step 2103. Otherwise, the processing after Step 1102 of the first embodiment is executed.
(Step 2103) When the write destination of the write process is the portable storage device 40, the process proceeds to Step 2104. Otherwise, go to Step 2105.
(Step 2104) The drive use processing calculation program 656 calculates the time required for Write from the data size to be written, the read performance of the logical volume, the write destination disk performance, and the load time. As with Step 1104, the time calculation method may be simply calculated from the data size, drive, logical volume performance, and load time, or the past performance history is saved and the processing time is improved by learning. May be. In the case of using a plurality of drives and a plurality of disks, the prefetch time may be shortened when processing in parallel is possible. Furthermore, you may shorten the part of the process which overlaps. Moreover, you may calculate by another means.
(Step 2105) The drive use processing calculation program 656 calculates the time required for Write from the data size to be written and the write logical volume performance.

With reference to FIG. 21, a job execution order and a method of determining a drive to be used will be described.
(Step 2201) The drive utilization processing calculation program 656 checks job information occupying the drive from the job correspondence management table 660.
(Step 2202) When the content of the job process detected in Step 2100 is the Write process, the process proceeds to Step 2203. Otherwise, the processing after Step 1202 of the first embodiment is executed.
(Step 2203) When the write destination of the write process is the portable storage device 40, the process proceeds to Step 2104. Otherwise, go to Step 2105.
(Step 2204) The drive use processing calculation program 656 searches the job correspondence management table 660 for a drive free time based on the write job execution time.
(Step 2205) If there is free time, the process proceeds to Step 2206. If there is no free time, the drive utilization processing calculation program 656 returns an error to the storage management program 654 or the job management program 604. As a result, an error can be displayed on the input / output device 670 or 620.
(Step 2206) The drive use processing calculation program 656 checks whether there is a job using the same drive and disk, and whether there is sufficient free time thereafter. If there is time, go to Step 2207. If there is no time, the process proceeds to Step 2208.
(Step 2207) The drive use processing calculation program 656 selects the same drive, and updates the start and end times of the job correspondence management table 660 so that they are immediately after the job using the same drive / disk. The drive use job column is blank.
(Step 2208) The drive use processing calculation program 656 selects a free drive.

  There are several methods as a drive selection method and a job order determination method. In addition to the method of Step 1206 of the first embodiment, for example, a method considering the shelf life can be mentioned. Among optical discs, a hologram disc has a life until data writing managed by the medium usage limit 6586 of the configuration management table 658, and is called a shelf life. In the case of a write job, if there is free capacity in the logical volume of the block storage, the execution time can be delayed, while it is necessary to execute the write job in consideration of this shelf life.

  Therefore, the capacity of the hologram disk can be used effectively by setting a Write job near a job whose shelf life is close and whose logical volume is empty. In addition, a write instruction for a portable storage medium may be received from a host or the like. This is performed with the intention of confirming that data has been stored in a non-rewritable storage medium. A write job based on such a write instruction is set in time for the instruction from the computer.

In addition, there is a case where the capacity of the hologram disc is wasted when the additional recording process is executed, and it is desirable to write the data as much as possible. For this reason, the write data is once stored in the HDD, the data is accumulated to some extent, and then the data is written to the hologram disk. As a method of collecting, a plurality of data having some relation may be collected. After selecting a drive that meets the deadline, the start and end times of the job correspondence management table 660 are updated, and the job processing of the upper AP is left blank.
(Step 2209) The storage management program 654 reduces the HDD free capacity after the occurrence of Write from the HDD free capacity table.

  With the above processing, it is possible to schedule the drive use processing of the portable storage device even in a system in which not only the read processing of the first embodiment but also the read / write processing coexists, and the wait of the read / write processing is less likely to occur. As a result, it is difficult for the upper AP to wait for processing.

  In the first and second embodiments, the prediction of the HDD capacity management table 661 of the block storage 31 is managed only by the processing in the hybrid storage device 91. In the third embodiment, not only the processing in the hybrid storage device 91 but also the data migration processing from the file server 20 and the block storage 30 in the upper hierarchy is used to predict the future capacity and execute the processing. Only differences from the second embodiment will be described below. The difference from the second embodiment is the sequence of information acquisition processing for each component (corresponding to Step 1030 in the first embodiment).

The information acquisition process Step 3030 according to the third embodiment will be described with reference to FIG.
(Step 3031) The system management program 655 of the device management server 65 periodically acquires information from the file server 20, block storages 30, 31, hybrid control server 50, portable storage device 40, job management server 60, and archive management server 70. .
(Step 3032) The system management program 655 blocks from the block storage 30 based on the last access time information of the data management table 706 acquired from the archive management server 70 and the movement policy information of the policy management table 705 in addition to the processing of Step 1032 of the first embodiment. Data to be moved to the storage 31 and its time are predicted.
(Step 3033) The storage management program 654 updates the information in the HDD capacity management table 661 from the data capacity and time to be moved. Furthermore, the table described in Step 1033 of the first embodiment is also updated.
With the above processing, in addition to the processing of the second embodiment, it is possible to schedule the drive use processing of the portable storage device in consideration of the data movement outside the hybrid storage device, and in the hierarchical storage system including the hybrid storage device As a result, waiting for Read / Write processing is less likely to occur, and as a result, waiting for processing of higher-level APs is less likely to occur.

  Up to the third embodiment, a search for free time is performed when determining the job execution order and the drive to be used, but if there is no free time, an error has occurred. In the fourth embodiment, when there is no free time, a process for checking whether or not a newly registered drive usage process is possible by shifting the job execution time of the already registered drive usage process is performed. Only differences from the first embodiment will be described below. The difference from the first embodiment is a sequence of job execution order determination and used drive determination (corresponding to Step 1200 of the first embodiment).

The job execution order determination and the used drive determination process Step 4200 according to the fourth embodiment will be described with reference to FIG.
(Step 4201) The processing from Step 1201 to Step 1203 of the first embodiment is executed.
(Step 4202) The drive use processing calculation program 656 searches the job correspondence management table 660 for a drive free time. If there is a time with sufficient free space, the process proceeds to Step 1206 of the first embodiment. If not, the process proceeds to Step 4203.
(Step 4203) The drive use processing calculation program 656 checks whether or not a free time is available if the drive and time of the already registered drive use job processing are changed. If free time is available, the process proceeds to Step 4204. If not, the drive utilization processing calculation program 656 returns an error to the storage management program 654 or the job management program 604. As a result, an error can be displayed on the input / output device 670 or 620.
(Step 4204) The drive use processing calculation program 656 changes the start time and end time of the job already registered in the job correspondence management table 660. There are several methods for this change method. For example, there is a method of canceling all registered drive utilization processing jobs and registering the job with the top AP starting from the job with the highest AP job start time or the longest processing time. Another example is processing for setting whether or not a new job can be registered by shifting the execution time from the last registered drive use job.

Or, if there are jobs that can be divided into jobs that use multiple drives or multiple data among registered jobs, search for those that can be divided, and if there are jobs that can be divided, subdivide the jobs and shift the time to free up time. You can also make it easier to create a job and check if the job can be set. In addition, since Read and Write are more likely to be urgent, Read is simply read from a computer such as a host except for a Write based on a Write instruction to the portable storage medium. You may prioritize. Also, the job start time and end time may be changed by other methods.
(Step 4205) The drive use processing calculation program 656 selects a free time and a drive, and determines the order of executing the drive use processing job at that time.
(Step 4206) The drive usage processing calculation program 656 creates and changes a script for executing the changed and added drive usage processing job.
(Step 4207) The system management program 655 registers a job for executing the created script and registers it in the job management program 604 so as to change the job for executing the changed script.

  As a result of the above processing, even if there is no drive available due to the already registered drive usage processing, it is possible to schedule new processing by changing the existing drive usage processing execution time, and wait for Read / Write processing. Can be difficult.

  Up to the fourth embodiment, it is assumed that the optical disk is present in the portable storage device. However, the optical disk is not necessarily present in the portable storage device. For example, to reduce power consumption, the optical disk is not included in the portable storage device. To go offline. In the fifth embodiment, a process for scheduling the use of a drive when there is an offline optical disk will be described. Only differences from the first embodiment will be described below. The difference from the first embodiment is that a table for offline management is added, and the sequence of time prediction processing for drive utilization processing (corresponding to Step 1100 of the first embodiment) is changed.

  FIG. 24A shows a portable storage device configuration management table 413B used for offline management. The portable storage device configuration management table 413B is obtained by adding an online / offline column 4135 to the portable storage device configuration management table 413 of the first embodiment. The offline portable storage medium management table 414 has the following information (4140) to (4143). The offline management table 414 exists both in the memory 403 in the portable storage device 40 and in the memory 653 in the device management server 65. The original data is in the memory 403 and is collected periodically by the storage management program 654. The offline information is updated by the portable storage device management program 412.

First, only the differences in the portable storage device configuration management table 413B will be described.
(4135) Online / offline. Indicates whether the disk is online or offline.

The offline portable storage medium management table 414 will be described with reference to FIG. 24B.
(4140) A portable storage medium identifier. It has information that uniquely identifies a portable storage medium.
(4141) Storage location. It has information on the identifier of the place where the portable storage medium is stored. This information can include any physical location (address, building, warehouse, shelf, etc.).
(4142) Tag. It has tag information for managing portable storage media. This information may be the same as the portable storage medium identifier or may be newly set for management.
(4143) Time required for acquisition. Have time to bring portable storage media online. This information may be accurate or approximate information.

FIG. 25 shows step 5100 of drive utilization processing time prediction when the portable storage medium is offline.
(Step 5101) The processing of Step 1101 and Step 1102 of the first embodiment is executed.
(Step 5102) The drive use processing calculation program 656 checks whether the position of the target data of the drive use processing is the portable storage device 40. In the case of the portable storage device 40, the process proceeds to Step 5105. If it is not the portable storage device 40, the process proceeds to Step 5103.
(Step 5103) The drive use processing calculation program 656 checks whether the position of the target data for the drive use processing is offline. If it is offline, the process proceeds to Step 5104. Otherwise, it is determined that the position of the target data is in the HDD, and the process proceeds to Step 5107.
(Step 5104) The drive use processing calculation program 656 acquires the time required for acquiring the portable storage medium from the offline portable storage medium management table 414.
(Step 5105) The drive use processing calculation program 656 acquires the read / write performance and load time of the drive and HDD from the configuration management table.
(Step 5106) Step 1105 of the first embodiment is executed.
(Step 5107) The drive utilization processing calculation program 656 determines that prefetching is unnecessary.

  With the above processing, even when the portable storage medium is outside the portable storage device 40 and is offline, it is possible to schedule the processing using the drive, and it is possible to make it difficult to wait for Read / Write processing.

Computer system: 1, Host computer: 10, Device management server: 65, Hybrid storage device 91

Claims (12)

A computer comprising a processor,
The processor is
A computer, wherein a prefetch schedule for copying data from a first storage medium to a second storage medium in advance is determined based on the time taken for the prefetch before a read request from an external computer. The computer according to claim 1,
A memory that stores a drive schedule, which is a schedule for loading the plurality of first storage media, in a drive capable of reading data from the plurality of first storage media;
The computer, wherein the processor determines the prefetch schedule based on the drive schedule. A computer according to claim 2, wherein
In the memory, application execution schedule information, data information used in the application, position information of data stored in the plurality of first storage media, information on the second storage medium, Is stored,
Based on the execution schedule information of the application, data information used in the application, location information of the stored data, information on the second storage medium, and the drive schedule, the prefetch schedule is determined. And determining and reflecting the determined drive use schedule in prefetching as a drive use job in the drive schedule. The computer according to claim 3, wherein
The processor is
Based on the application execution schedule, the location information of the stored data, and the time taken for the prefetch, calculate the load start time at which the prefetch of the data used by the application can be completed before the application execution starts,
Referring to the drive schedule, check whether the drive at the load start time is free,
When the drive is not free at the load start time, the prefetch schedule is determined by advancing the prefetch time of data used by the application or the prefetch scheduled to use the drive at the time. calculator. The computer according to claim 4, wherein
The processor is
Based on the execution schedule information of the application, the data used by the application is deleted from the second storage medium of the application at the execution completion time of the application,
Based on the information in the second storage medium,
Determining the prefetch schedule so that there is no time for the amount of data stored in the second storage medium to exceed a threshold;
A computer characterized by that. The computer according to claim 3, wherein
Connected to a first storage device comprising the first storage medium and a second storage device comprising the second storage medium;
Based on the determined prefetch schedule, a read instruction is transmitted to the first storage, and data is stored in the second storage medium from the first storage,
The computer, wherein the external computer connected to the second storage device accesses data stored in the second storage medium. The computer according to claim 6, wherein
The drive has a function of writing data to the first storage medium,
The processor determines the drive schedule in consideration of data movement from the second storage medium to the first storage medium;
A computer characterized by that. The computer according to claim 7, wherein
The memory further stores a data storage start time limit of the first storage medium,
The computer, wherein the processor determines the drive schedule based on a first storage medium data storage start time limit. The computer according to claim 5, wherein
From the first storage device comprising the first storage medium, the second storage device comprising the second storage medium, the third storage device comprising a third storage medium, and the third storage device Connected to an archive management server that manages movement to the second storage device,
The processor acquires a last access time and a tier migration policy of management information of the third storage medium storage data from the third storage device or the archive management server, and based on the acquired information, the third storage policy Predicting data and time to move from the storage device to the second storage device,
The computer updates the information in the second storage medium based on the prediction. The computer according to claim 5, wherein
The processor is
Refer to the drive schedule,
When there is a drive use job that reads the same data as the data used in the application, the drive use job that loads the same first storage medium is executed based on the information of the second storage medium When the data used in the application is stored in the second storage medium until the execution completion time of the application, it is confirmed whether the amount of data stored in the second storage medium exceeds the threshold And
If there is no time for the stored data amount of the second storage medium to exceed the threshold value, the drive use job that reads the same data is used as the related drive use without adding the drive use job to the drive schedule at the time of prefetching the application. A computer that is managed as a job. The computer according to claim 10, wherein
If the processor receives an application completion notification,
Refer to execution schedule information of the application and data information used in the application, determine whether data used for execution of the application is managed as a related drive use job of another application,
A computer that deletes data used for execution of the application from the second storage medium when it is not managed as a related drive use job of another application. A first storage device comprising a plurality of portable storage media; a second storage device comprising a prefetch storage medium capable of reading data at a higher speed than the portable storage media; the first storage device; A hybrid management server that provides a hybrid storage system from the second storage device to the host computer and a prefetch for copying data from the first storage device to the second storage device in advance before a read request from the host computer is managed. A storage system comprising a device management server
The first storage device includes a smaller number of drives for reading data from the plurality of portable storage media than the plurality of portable storage media,
The device management server includes a drive schedule that is a schedule for loading the plurality of portable storage media, information on the prefetch storage medium, execution schedule information of a host application, and data stored in the plurality of portable storage media. A storage system, wherein the prefetch schedule is determined based on position information.

Images