WO2014010016A1

WO2014010016A1 - Program, data management method, and information processing device

Info

Publication number: WO2014010016A1
Application number: PCT/JP2012/067478
Authority: WO
Inventors: 英治浜本; 修三西野; 尚小山内; 秀行安本; 藤田　大輔; 輝中田; 幸美橋本; 山上　正人; 正勝岩本; 英季堀岡; 俊介元井
Original assignee: 富士通株式会社
Priority date: 2012-07-09
Filing date: 2012-07-09
Publication date: 2014-01-16

Abstract

The present invention improves data-reading efficiency. Storing means (1a, 1b, 1c, 1d, 1e) are associated, respectively, with a plurality of time periods (T1, T2, T3, T4, T5). Storing means (1a, 1b, 1c, 1d, 1e) store data updated within the corresponding time periods. Upon receiving input of a read request for data (5), a determination means (2) determines, in a prescribed order, whether the data (5) is stored in each of the storing means. A reading means (3) reads the data (5) from the storing means (1a) in which the data (5) designated by the read request is stored. A storage means (4) then stores the data (5) in a storing means (1d), which has an earlier prescribed order than the storing means (1a) does.

Description

Program, data management method, and information processing apparatus

The present invention relates to a program for managing data stored in a plurality of storage devices, a data management method, and an information processing apparatus.

There is a technology called snapshot as a storage device management technology. Snapshot is saving the contents of a disc at a certain point in a reproducible state. When creating a snapshot of a disk in operation, for example, subsequent writing to the disk is prohibited and the disk is stored as an original disk. Another disk is used as a new operation disk, and difference data from the time of snapshot creation is stored in the new operation disk.

Snap shot creation can be repeated. In the creation of the second and subsequent snapshots, subsequent writing to the active disk at that time is prohibited, and a new active disk for storing subsequent differential data is added. As described above, when snapshots are repeatedly collected, the disk for storing the difference data has a multistage configuration.

When reading data after creating a snapshot, the data is read from the saved original disk or the disk storing the difference data. If the disk storing the difference data has a multi-stage configuration, for example, it is determined in order from the disk storing the new difference data whether or not the data to be read is stored. If no corresponding data is stored in any disk that stores the difference data, it is finally determined whether or not the original disk has the data. Then, the data to be read is read from the disk storing the data to be read.

For example, when copying data, a technology that copies only the data recorded in the data area to be written to the copy destination disk without copying all the data recorded on the copy source disk to the copy destination disk. There is. In this technique, when the data to be read does not exist on the copy destination disk, the data is retrieved from the copy source disk.

JP 2007-172082 A

However, in the method of adding a disk for storing differential data each time a snapshot is created, the data read efficiency deteriorates as the number of snapshot creation increases. For example, if data exists on the original disk in data reading after repeated creation of snapshots, the presence / absence of data to be read is determined for a large number of disks. Therefore, when snapshot creation is repeated, it takes time to read data existing on the original disk or a disk that was used as an operational disk in the early days.

In one aspect, an object of the present invention is to provide a program, a data management method, and an information processing apparatus that improve data reading efficiency.

In one aspect of the program, storage means is associated with each of a plurality of time zones, and the storage means stores data updated within the corresponding time zone, and receives a data read request. And determining whether or not the data is stored in a predetermined order for each storage means, and reading the data from the storage means storing the data specified by the read request, than the storage means The information processing apparatus is caused to execute the process of storing the data in the storage means having a predetermined order earlier.

According to one aspect, data read efficiency is improved.
These and other objects, features and advantages of the present invention will become apparent from the following description taken in conjunction with the accompanying drawings which illustrate preferred embodiments by way of example of the present invention.

It is a figure which shows the function structural example of the information processing apparatus which concerns on 1st Embodiment. It is a figure which shows the system configuration example of 2nd Embodiment. It is a figure which shows one structural example of the hardware of the management server used for this Embodiment. It is a block diagram which shows the function of a management server. It is a figure which shows an example of the arrangement | sequence and internal structure of a virtual disk. It is a figure which shows an example of the data structure of a management information storage part. It is the first half of the flowchart which shows an example of a data read-out process. It is the latter half of the flowchart which shows an example of a data read-out process. It is a figure which shows the example of the read-out process of the data from the operation | use disk. It is a figure which shows the initial value of the movement number of the read data. It is a figure which shows the update example of the total number of movements based on the read-out process of the data from an operation disk. It is a figure which shows the example of the 2nd read-out process of data. It is a figure which shows the update example of the movement total according to the reading of the data of the 2nd time. It is a figure which shows the judgment condition of the free capacity of the data copy destination. It is a figure which shows the example of a copy of data. It is a figure which shows the example of the read-out process of the 3rd time of data. It is a figure which shows the example of the read-out process of the data of a snapshot. It is a figure which shows the example of a setting of the total number of movement at the time of the data reading of a snapshot. It is a figure which shows the example of the reading process of the 2nd time of the data of a snapshot. It is a figure which shows the example of the data copy according to reading of the data of a snapshot.

Hereinafter, the present embodiment will be described with reference to the drawings. Each embodiment can be implemented by combining a plurality of embodiments within a consistent range.
[First Embodiment]
First, a first embodiment will be described. In the first embodiment, in an environment in which storage means for storing difference data by snapshots has a multi-stage configuration, at the time of reading data, the determination that the absence of data is performed earlier than the storage means of the reading source is performed. The means stores the read data. As a result, the time for reading data is shortened by reducing the number of storage means for determining the presence / absence of data when reading the same data thereafter.

FIG. 1 is a diagram illustrating a functional configuration example of the information processing apparatus according to the first embodiment. The information processing apparatus X according to the first embodiment includes a plurality of

storage units

1a, 1b, 1c, 1d, and 1e, a determination unit 2, a reading unit 3, and a storage unit 4.

The storage means 1a, 1b, 1c, 1d, and 1e are associated with a plurality of time zones T1, T2, T3, T4, and T5, respectively. The plurality of time zones T1, T2, T3, T4, and T5 are time zones that are divided using, for example, snapshot creation request input times t1, t2, t3, and t4 as boundaries. The storage means 1a, 1b, 1c, 1d, and 1e store data updated within the corresponding time zone.

Of the

storage units

1a, 1b, 1c, and 1d, the

storage units

1a, 1b, 1c, and 1d corresponding to the time zones T1, T2, T3, and T4 before the last snapshot creation request input time t4 are Writing data in response to an external write request is prohibited. Thereby, data at the time of inputting the snapshot creation request is held. On the other hand, the storage unit 1e corresponding to the time period T5 after the input time t4 of the last snapshot creation request is a storage unit in operation, and stores data in response to an external data write request. The

When the data read request is input, the determination unit 2 sequentially determines whether or not the data is stored in the storage unit according to a predetermined determination order. For example, the determination unit 2 sequentially selects the corresponding time zone from the later storage unit, and determines whether the data to be read is stored in the selected storage unit. In this case, selection and determination are performed in the order of the storage unit 1e, the storage unit 1d, the storage unit 1c, the storage unit 1b, and the storage unit 1a.

Note that a request to read data held in a snapshot may be input. In this case, the determination unit 2 stores, for example, data to be read according to a predetermined determination order for the storage unit corresponding to the time zone before the input time point of the snapshot creation request instructing the creation of the snapshot. It is determined in turn whether or not.

Read means 3 reads the data 5 from the storage means storing the data specified by the read request. For example, the reading unit 3 outputs the read data 5 as a response to the read request. The reading unit 3 temporarily stores, for example, the read data 5 as a cache.

The storage unit 4 stores the data 5 in a storage unit that is earlier in the determination order than the storage unit in which the data 5 is stored. For example, the storage unit 4 receives the data 5 held as a cache from the reading unit 3 and stores it in any storage unit. For example, the storage unit 4 stores the data in any of the

storage units

1a, 1b, 1c, and 1d corresponding to the time zone before the input time point of the last snapshot creation request.

In the storage means 4, for example, the data storage conditions are defined in advance, and the data storage conditions are determined for each storage means that has a predetermined order earlier than the storage means in which the data 5 designated by the read request is stored. Determine if it is satisfied. Then, the storage unit 4 stores the read data 5 in the storage unit that satisfies the write condition.

The storage unit 4 includes, for example, a storage unit that stores data for each storage unit that has an earlier determination order than the storage unit that stores the read data each time data is read. The value which shows the difference of the order of judgment between is calculated | required. Then, the storage unit 4 adds the value obtained for each storage unit to a determination value indicating whether the storage unit needs to store the read data. In this case, the storage unit 4 stores the read data in the storage unit whose determination value exceeds a predetermined threshold value.

According to such an information processing apparatus X, when a data read request is input, whether or not data is stored in the

storage units

1a, 1b, 1c, 1d, and 1e by the determination unit 2 is determined in advance. The determination is made in order according to the determination order. Next, the reading unit 3 reads the data from the storage unit storing the data specified by the reading request. Then, the storage means 4 stores the read data in the storage means that is earlier in the determination order than the read storage means.

Thereby, when the same data read request is input next, the data can be read from the storage means whose judgment order is earlier than the previous time. That is, it is possible to reduce the number of storage means to be determined as to whether or not data to be read is stored. As a result, the data read efficiency is improved.

For example, in the example of FIG. 1, it is assumed that the data 5 specified by the read request is stored in the last storage unit 1a as the order of determination. In this case, the determination unit 2 sequentially determines the presence / absence of data from the storage unit 1e. Then, the determination unit 2 determines that the data 5 is stored in the fifth storage unit 1a in the determination order.

The reading unit 3 receives the identification information of the storage unit 1a in which the data 5 is stored from the determination unit 2, and reads the data 5 from the storage unit 1a. The reading unit 3 outputs the read data 5 as a response to the read request.

The storage unit 4 receives the data 5 from the reading unit 3 and stores the data 5 in any of the storage units whose determination order is earlier than that of the storage unit 1a. For example, the storage unit 4 stores the data 5 in the storage unit 1d corresponding to the latest time zone among the storage units corresponding to the time zone before the input time t4 of the last snapshot creation request. As a result, the data 5 is stored in the storage unit 1d having the earliest determination order among the storage units that are prohibited from writing according to a write request from the outside by the snapshot.

Thus, when there is a next request to read data 5, data 5 can be read from storage means 1d. Since the storage unit 1d precedes the storage unit 1a as the determination order of the presence / absence of data, the number of storage units that determine the presence / absence of data when reading the data 5 is small. As a result, the data read efficiency is improved.

In addition, since the storage means storing the data is distributed in plural, the I / O load of each storage means is reduced. In addition, reliability is improved by reducing the reading of data that goes back many steps in the hierarchy. Furthermore, since the data is copied to the free area of the storage means that is prohibited from writing data in response to an external write request, the free area can be used effectively, and the resource utilization efficiency is improved.

Note that the determination unit 2, the reading unit 3, and the storage unit 4 can be realized by a processor included in the information processing apparatus X, for example. The

storage units

1a, 1b, 1c, 1d, and 1e can be realized by, for example, a RAM (Random Access Memory) included in the information processing apparatus X.

Also, the lines connecting the elements shown in FIG. 1 indicate a part of the communication path, and communication paths other than the illustrated communication paths can be set.
[Second Embodiment]
Next, a second embodiment will be described. In the second embodiment, data is efficiently read from a snapshot created using a virtual disk that is a combination of a plurality of RAID (Redundant Array of Inexpensive Disks) devices.

FIG. 2 is a diagram illustrating a system configuration example according to the second embodiment. The management server 100 is connected to the two

networks

21 and 22. A plurality of

business servers

31, 32,... Are connected to the network 21. A plurality of

RAID devices

41, 42, 43,... Are connected to the network 22.

The management server 100 is a computer that creates virtual disks using

RAID devices

41, 42, 43,. For example, the management server 100 uses a software RAID technology and generates a virtual RAID group by combining LUs (Logical Units) of

RAID devices

41, 42, 43,. The management server 100 sets the virtual RAID group generated in this way as a virtual disk.

Management server 100 constructs a logical volume with one or a plurality of virtual disks. Then, the management server 100 accepts access specifying the address in the logical volume from the

business servers

31, 32,. Then, the management server 100 accesses the physical area in the RAID device assigned to the designated address.

In addition, the management server 100 can create a snapshot of information stored in the logical volume. For example, when the management server 100 receives a snapshot creation request from the

business servers

31, 32,..., The subsequent access to the data stored in the logical volume at that time is limited to read only (read only). To do. Thereafter, when there is a request to write data to the logical volume, the management server 100 stores the data as difference data.

FIG. 3 is a diagram illustrating a configuration example of hardware of the management server used in the present embodiment.
The management server 100 is entirely controlled by a processor 101. A RAM 102 and a plurality of peripheral devices are connected to the processor 101 via a bus 109. The processor 101 may be a multiprocessor. The processor 101 is, for example, a central processing unit (CPU), a micro processing unit (MPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC), or a programmable logic device (PLD). The processor 101 may be a combination of two or more elements among CPU, MPU, DSP, ASIC, and PLD.

The RAM 102 is used as a main storage device of the management server 100. The RAM 102 temporarily stores at least part of an OS (Operating System) program and application programs to be executed by the processor 101. The RAM 102 stores various data necessary for processing by the processor 101.

Peripheral devices connected to the bus 109 include an HDD (Hard Disk Drive) 103, a graphic processing device 104, an input interface 105, an optical drive device 106, a device connection interface 107, and

network interfaces

108a and 108b.

The HDD 103 magnetically writes and reads data to and from the built-in disk. The HDD 103 is used as an auxiliary storage device of the management server 100. The HDD 103 stores an OS program, application programs, and various data. Note that a semiconductor storage device such as a flash memory can also be used as the auxiliary storage device.

A monitor 11 is connected to the graphic processing device 104. The graphic processing device 104 displays an image on the screen of the monitor 11 in accordance with a command from the processor 101. Examples of the monitor 11 include a display device using a CRT (Cathode Ray Tube) and a liquid crystal display device.

A keyboard 12 and a mouse 13 are connected to the input interface 105. The input interface 105 transmits a signal transmitted from the keyboard 12 or the mouse 13 to the processor 101. The mouse 13 is an example of a pointing device, and other pointing devices can also be used. Examples of other pointing devices include a touch panel, a tablet, a touch pad, and a trackball.

The optical drive device 106 reads data recorded on the optical disk 14 using a laser beam or the like. The optical disk 14 is a portable recording medium on which data is recorded so that it can be read by reflection of light. The optical disk 14 includes a DVD (Digital Versatile Disc), a DVD-RAM, a CD-ROM (Compact Disc Read Only Memory), a CD-R (Recordable) / RW (ReWritable), and the like.

The device connection interface 107 is a communication interface for connecting peripheral devices to the management server 100. For example, the memory device 15 and the memory reader / writer 16 can be connected to the device connection interface 107. The memory device 15 is a recording medium equipped with a communication function with the device connection interface 107. The memory reader / writer 16 is a device that writes data to the memory card 17 or reads data from the memory card 17. The memory card 17 is a card type recording medium.

The network interface 108a is connected to the network 21. The network interface 108a transmits / receives data to / from the

business servers

31, 32,... Via the network 21.

The network interface 108b is connected to the network 22. The network interface 108b transmits / receives data to / from the

RAID devices

41, 42, 43,.

With the hardware configuration as described above, the processing function of the second embodiment can be realized. The information processing apparatus X shown in the first embodiment can also be realized by hardware similar to the management server 100 shown in FIG.

The management server 100 implements the processing functions of the second embodiment by executing a program recorded on a computer-readable recording medium, for example. The program describing the processing contents to be executed by the management server 100 can be recorded in various recording media. For example, a program to be executed by the management server 100 can be stored in the HDD 103. The processor 101 loads at least a part of the program in the HDD 103 into the RAM 102 and executes the program. A program to be executed by the management server 100 can also be recorded on a portable recording medium such as the optical disc 14, the memory device 15, and the memory card 17. The program stored in the portable recording medium becomes executable after being installed in the HDD 103 under the control of the processor 101, for example. The processor 101 can also read and execute a program directly from a portable recording medium.

Next, the data access and snapshot creation function in the management server 100 will be described.
FIG. 4 is a block diagram illustrating functions of the management server. The management server 100 includes a data management unit 110, a management information storage unit 120, a snapshot creation unit 130, and an access unit 140.

The data management unit 110 manages the logical volume 110a using the

virtual disks

111, 112,. The

virtual disks

111, 112,... Are, for example, RAID groups (or LUs within the RAID group) created by software RAID using LUs of

RAID devices

41, 42, 43,.

The management information storage unit 120 stores management information used for creating a snapshot and accessing data. Details of the management information will be described later (see FIG. 6).
The snapshot creation unit 130 creates a snapshot of the logical volume 110a managed by the data management unit 110. For example, the snapshot creation unit 130 creates a snapshot in response to a snapshot creation request from the

business servers

31, 32,. When creating a snapshot, the snapshot creation unit 130 causes the data management unit 110 to create a new virtual disk and sets the virtual disk as a new operational disk. Thereafter, data writing to the logical volume 110a is performed on the new operation disk. As a result, the data state at the time of creating the snapshot is stored in the virtual disk incorporated in the logical volume 110a before the creation of the snapshot.

The access unit 140 accesses data in the logical volume 110a managed by the data management unit 110 in response to a data access request from the

business servers

31, 32,. For example, the access unit 140 writes data to the operation disk. Further, when reading data, the access unit 140 executes a data read process in response to the read request while tracing the virtual disk in order from the active disk. In the process of reading data to each virtual disk, the access unit 140 refers to the update block list corresponding to the virtual disk and determines whether or not a block including the data to be read is stored in the virtual disk. To do. When the data to be read is not stored in the virtual disk, the access unit 140 sequentially executes a read process to the next virtual disk. Then, the access unit 140 reads a block including the corresponding data from the virtual disk having the data to be read.

In addition, when the read data satisfies a predetermined condition, the access unit 140 copies a block including the data to a virtual disk whose determination order of presence / absence of data is earlier. The access unit 140 can copy data only when it is determined that it is useful for improving the efficiency of the reading process. For example, the access unit 140 copies a block including data that greatly contributes to the efficiency of data reading in consideration of the number of times data is read and the number of virtual disks that determine the presence or absence of data. To do.

The access unit 140 is an example of a function that includes the determination unit 2, the reading unit 3, and the storage unit 4 in the first embodiment shown in FIG. The

virtual disks

111, 112,... Are examples of the

storage units

1a, 1b, 1c, 1d, and 1e in the first embodiment shown in FIG. Furthermore, the lines connecting the elements shown in FIG. 4 indicate a part of the communication paths, and communication paths other than the illustrated communication paths can be set.

Next, the arrangement and internal structure of the virtual disk will be described.
FIG. 5 is a diagram showing an example of the arrangement and internal structure of a virtual disk. In the example of FIG. 5, five virtual disks 111 to 115 are included in the logical volume 110a. The virtual disks 111 to 115 are arranged from the left in the order of early use as the operation disks. Data is written in response to a write request from the

business servers

31, 32,... Reading data in response to a read request from the

business servers

31, 32,... Is executed in order from a virtual disk that has been used as a working disk later. In the arrangement of FIG. 5, data reading in response to the read request is sequentially performed from the rightmost virtual disk 115 sequentially to the left.

The virtual disk 111 is an original disk that stores data (original data) written before a snapshot is created once. The virtual disk 111 is prohibited from writing data based on a write request from the

business servers

31, 32,... After the first snapshot (snapshot # 1) is created. The original data stored in the virtual disk 111 corresponds to the first snapshot.

The virtual disks 111 to 115 are arranged in order from the earliest used virtual disk with the virtual disk 111 as the original disk at the head. In the example of FIG. 5, the virtual disk 111 is first, the virtual disk 112 is second, the virtual disk 113 is third, the virtual disk 114 is fourth, and the virtual disk 115 is fifth.

The virtual disk 112 stores data written between the creation of the first snapshot and the creation of the second snapshot (snapshot # 2). Data written to the virtual disk 112 is difference data from the original data. The virtual disk 112 is prohibited from writing data based on a write request from the

business servers

31, 32,... After the second snapshot is created. The data stored in the

virtual disks

111 and 112 corresponds to the second snapshot.

The virtual disk 113 stores data written between the time when the second snapshot is created and the time when the third snapshot (snapshot # 3) is created. Data written to the virtual disk 113 is difference data from the second snapshot. The virtual disk 113 is prohibited from writing data based on a write request from the

business servers

31, 32,... After the third snapshot creation. The data stored in the virtual disks 111 to 113 corresponds to the third snapshot.

The virtual disk 114 stores data written between the third snapshot creation and the fourth snapshot (snapshot # 4) creation. Data written to the virtual disk 114 is differential data from the third snapshot. The virtual disk 114 is prohibited from writing data based on a write request from the

business servers

31, 32,... After the fourth snapshot is created. The data stored in the virtual disks 111 to 114 corresponds to the fourth snapshot.

The virtual disk 115 stores data written after the fourth snapshot creation. Data written to the virtual disk 115 is difference data from the fourth snapshot. The virtual disk 115 is still used as an operation disk and is writable.

When creating the virtual disks 111 to 115, the data management unit 110 allocates a predetermined amount (for example, 2 GB) of storage area to the created virtual disks. When data is written to the operating virtual disk and the free space becomes a predetermined threshold (for example, 800 MB) or less, the data management unit 110 adds a predetermined amount (for example, 2 GB) of storage area to the virtual disk. In this way, a certain amount of free space is always secured in the operating virtual disk. As a result, even if a large amount of data is written, it is not necessary to deplete the storage capacity of the operation disk.

However, by performing such an operation, each of the virtual disks 111 to 115 includes free space. Therefore, in the second embodiment, the free areas of the virtual disks 111 to 114 in which data writing based on the write request from the

business servers

31, 32,... Is prohibited are used as the copy destination of the read data. In this way, the free space is effectively utilized. Whether to copy the data to be read is determined based on the total number of movements of the data. The total number of movements is a cumulative value of the number of movements obtained each time data is read. The number of movements is a value indicating a difference in arrangement in the logical volume 110a from the virtual disk in which the read data is stored to the virtual disk corresponding to the read block management area. The total number of data movements is counted for each virtual volume and held in the management information storage unit 120. The total number of movements is an example of a determination value in the first embodiment.

Next, the management information stored in the management information storage unit 120 will be described in detail.
FIG. 6 is a diagram illustrating an example of the data structure of the management information storage unit. The management information storage unit 120 stores a plurality of read

block management areas

121a, 121b, 121c,... And a plurality of update block lists 122a, 122b, 122c,.

The read

block management areas

121a, 121b, 121c,... Are provided for each virtual disk newly incorporated in the logical volume 110a when the snapshot is created. For example, device numbers of corresponding virtual disks are set in the read

block management areas

121a, 121b, 121c,. The device number is unique information in the virtual disk system. In addition, the read

block management areas

121a, 121b, 121c,... Include numerical values that serve as indicators for whether or not the read data is to be copied to the corresponding virtual disk. For example, in the read

block management areas

121a, 121b, 121c,..., A set of the read data offset and the number of movements is registered. The offset is an address difference from the beginning of the storage area of the logical volume 110a to the data.

The update block lists 122a, 122b, 122c,... Are provided for each virtual disk newly incorporated in the logical volume 110a when the snapshot is created. For example, the device numbers of the corresponding virtual disks are set in the update block lists 122a, 122b, 122c,. In addition, offsets of blocks written in the corresponding virtual disks are registered in the update block lists 122a, 122b, 122c,. The block offset is, for example, an address difference from the beginning of the storage area of the logical volume 110a to the block.

Next, the procedure for reading data will be described. The data read process is executed when a data read request is input. In the data read request, the virtual disk that is the read start point can be specified. For example, when data is read from the logical volume 110a in operation, the virtual disk that is used as the operation disk at that time is designated as the read start point. When reading snapshot data created in the past, the virtual disk that was used as the active disk until immediately before the snapshot creation is designated as the starting point.

FIG. 7 is the first half of a flowchart showing an example of the data reading process. In the following, the process illustrated in FIG. 7 will be described in order of step number.
[Step S101] When a data read request is input, the access unit 140 selects a virtual disk as a starting point in the read request.

[Step S102] The access unit 140 determines whether there is data to be read on the selected virtual disk. For example, when the offset of a block including data to be read is registered in the update block list corresponding to the selected virtual disk, the access unit 140 determines that there is data to be read on the selected virtual disk. If there is data to be read, the access unit 140 proceeds with the process to step S104. If there is no data to be read, the access unit 140 proceeds with the process to step S103.

[Step S103] The access unit 140 newly selects a virtual disk that has been used as an active disk and advances the process to step S102. For example, the access unit 140 performs a data read process on the next virtual disk as a subroutine of a read process on the currently selected virtual disk.

[Step S104] When the data to be read is stored in the selected virtual disk, the access unit 140 reads a block including the data from the selected virtual disk. The access unit 140 transmits the read data to the business server that is the transmission source of the read request. The access unit 140 can also store the read data for each block including the data in a storage area managed by the access unit 140 in the RAM 102. Thereafter, the access unit 140 proceeds with the process to step S111 (see FIG. 8).

FIG. 8 is the second half of the flowchart showing an example of the data reading process. In the following, the process illustrated in FIG. 8 will be described in order of step number.
[Step S111] The access unit 140 sets 0 as the number of movements.

[Step S112] The access unit 140 selects a virtual disk that has been used as a working disk in the array by the next time. For example, the access unit 140 ends the reading process for the currently selected virtual disk, and returns to the process that has called the process. Then, the access unit 140 selects a virtual disk that has been a read target in the restored process.

[Step S113] The access unit 140 determines whether the selected virtual disk is a working disk. If the selected virtual disk is the active disk, the access unit 140 advances the process to step S117. If the selected virtual disk is not a working disk, the access unit 140 proceeds with the process to step S114.

[Step S114] If the selected virtual disk is not the active disk, the access unit 140 adds 1 to the number of movements.
[Step S115] The access unit 140 adds the current number of movements to the total number of movements of data to be read in the selected virtual disk. For example, the access unit 140 specifies the read block management area corresponding to the virtual disk based on the device number of the selected virtual disk from the management information storage unit 120. Next, the access unit 140 adds the current number of movements to the total number of movements corresponding to the offset value of the read data within the specified read block management area.

[Step S116] The access unit 140 determines whether the selected virtual disk is a virtual disk from which data is read. If the selected virtual disk is the read start point, the access unit 140 advances the process to step S117. If the selected virtual disk is not the read start point, the access unit 140 advances the process to step S112.

[Step S117] The access unit 140 determines whether there is a virtual disk in which the total number of data to be read exceeds the threshold. For example, the access unit 140 extracts the total number of moves corresponding to the offset value of the data to be read from all the read

block management areas

121a, 121b, 121c,... In the management information storage unit 120, and exceeds the threshold value. Judge whether there is a total number of moves. If there is at least one virtual disk whose total migration exceeds the threshold, the access unit 140 advances the process to step S118. If there is no virtual disk whose total migration exceeds the threshold, the access unit 140 ends the process without copying the data.

[Step S118] When there is a virtual disk whose total number of movements exceeds the threshold, the access unit 140 determines whether or not the virtual disk has enough free space to copy the data to be read. For example, the access unit 140 compares the free capacity of the virtual disk whose total migration exceeds the threshold with the data amount of the data to be read. If the free space is greater than or equal to the data amount, the access unit 140 determines that the virtual disk has sufficient free space. If there is sufficient free space, the access unit 140 proceeds with the process to step S119. If there is not enough free space, the access unit 140 ends the data reading process.

[Step S119] The access unit 140 copies the data to be read to a virtual disk in which the total number of movements of the data to be read exceeds a threshold and has sufficient free space. When there are a plurality of corresponding virtual disks, for example, the access unit 140 copies data to the last virtual disk in the array. When the block including the data is held in the RAM 102 when reading the data, the access unit 140 stores the block held in the RAM 102 in the virtual disk. Thereafter, the process ends.

Hereinafter, the data reading process will be described using a specific example. First, an example of data read processing starting from the operation disk will be described. In the second embodiment, the read request other than the read request for the snapshot data created in the past starts from the operation disk.

FIG. 9 is a diagram showing an example of data read processing starting from the operation disk. In FIG. 9, a data read request 51 stored in the virtual disk 111 which is the original disk is input. A block including data to be read is designated by an offset value “xxxxx”.

The access unit 140 determines whether or not data to be read is stored in order from the virtual disk 115 that is the active disk. Then, the access unit 140 detects that the data to be read is stored in the virtual disk 111 determined last, and reads the corresponding data 52 from the virtual disk 111 for each block including the data 52. The read data 52 is transmitted to the business server that transmitted the read request 51.

When the data 52 is read, the number of movements in each virtual disk for which the presence / absence of the data 52 has been determined is determined.
FIG. 10 is a diagram illustrating an initial value of the number of movements of read data. First, as the initial value of the movement number, the movement number of the virtual disk 111 from which the data 52 has been read is set to “0”. Thereafter, the number of movements is updated while tracing the virtual disk array up to the virtual disk 115 at the starting point of the read request 51. For example, each time the user moves to the next virtual disk, the number of movements is incremented by one.

FIG. 11 is a diagram showing an example of updating the total number of movements based on data read processing starting from the operation disk. The movement number of the virtual disk 111 from which the data 52 has been read is “0”, and the movement number of the next virtual disk 112 is “1”. Therefore, the total number of movements “1” is set in the read block management area 121 a corresponding to the virtual disk 112 in association with the offset value “xxxxx” of the block including the data 52.

The number of movements of the virtual disk 112 after the virtual disk 112 is “2”. Therefore, the total number of movements “2” is set in the read block management area 121 b corresponding to the virtual disk 113 in association with the offset value “xxxxx” of the block including the data 52.

The number of movements in the virtual disk 114 next to the virtual disk 113 is “3”. Accordingly, the total number of movements “3” is set in the read block management area 121 c corresponding to the virtual disk 114 in association with the offset value “xxxxx” of the block including the data 52.

In the second embodiment, the virtual disk 115 used as the operation disk is excluded from the copy destination of the read data. Therefore, the total number of movements corresponding to the virtual disk 115 is not updated.

When the update of the total number of moves in each read

block management area

121a, 121b, 121c is completed, it is determined whether there is a total number of moves exceeding the threshold. For example, it is assumed that the threshold of the total number of movements for copying data is “5”. Then, in the example shown in FIG. 11, the total number of movements corresponding to the block including the data 52 does not exceed 5 for all virtual disks.

After that, it is assumed that the data 52 is read a second time starting from the operation disk.
FIG. 12 is a diagram illustrating an example of the second data read process. When the second read request 53 for the data 52 is input, the presence or absence of the data 52 is determined in order from the virtual disk 115 that is the starting point, as in the previous case. Then, the data 52 is read from the virtual disk 111 storing the data 52. Then, in accordance with the second reading of the data 52, the total number of movements of the data 52 in each virtual disk is updated.

FIG. 13 is a diagram showing an example of updating the total number of movements according to the second data read. The number of movements in each of the virtual disks 111 to 114 is the same as in the first case shown in FIG.

The number of movements in the virtual disk 112 is “1”. Therefore, in the read block management area 121a corresponding to the virtual disk 112, “1” is added to the total number of movements corresponding to the offset value “xxxxx” of the block including the data 52 and updated to “2”.

The number of movements in the virtual disk 113 is “2”. Therefore, in the read block management area 121b corresponding to the virtual disk 113, “2” is added to the total number of movements corresponding to the offset value “xxxxx” of the block including the data 52 and updated to “4”.

The number of movements in the virtual disk 114 is “3”. Therefore, in the read block management area 121c corresponding to the virtual disk 114, “3” is added to the total number of movements corresponding to the offset value “xxxxx” of the block including the data 52 and updated to “6”.

In the example of FIG. 13, the total number of movements “6” corresponding to the data 52 in the virtual disk 114 exceeds the threshold value “5”. Therefore, it is determined whether or not the virtual disk 114 has enough free space to copy the data 52.

FIG. 14 is a diagram showing the determination status of the free capacity of the data copy destination. In the example of FIG. 14, the data amount of the data 52 stored in the virtual disk 111 is a byte (a is a positive real number), and the free space capacity of the virtual disk 114 is b bytes (b is a positive real number). . In this case, it is determined whether or not the condition “a ≦ b” is satisfied. In the example of FIG. 14, it is assumed that the condition is satisfied. In that case, the data 52 is copied to a free area of the virtual disk 114.

FIG. 15 is a diagram showing a copy example of data. As shown in FIG. 15, the data 52 of the virtual disk 111 is copied to the virtual disk 114. The copied data is stored in the virtual disk 114 as difference data. At this time, the offset value of the block including the data 52 is set in the update block list (see FIG. 6) corresponding to the virtual disk 114.

Thereafter, when the third read request for the data 52 is input starting from the operation disk, the data 52 is read from the virtual disk 114.
FIG. 16 is a diagram illustrating an example of the third data read process. When the third read request 54 for the data 52 is input, the presence / absence of the data 52 is determined in order from the virtual disk 115 as the starting point. Then, since the offset value of the block including the data 52 is set in the update block list of the virtual disk 114, it is determined that the data 52 is stored in the virtual disk 114. Therefore, the data 52 is read from the virtual disk 114. The fourth and subsequent readings of the data 52 can also be performed from the virtual disk 114.

In this way, when the total number of data movements exceeds the threshold value, the data can be quickly read out by copying the data to a virtual disk that has a fast determination order of the presence or absence of data at the time of reading. . In the examples of FIGS. 9 to 16, when the first and second data 52 are read, the presence / absence of the data 52 is determined for five virtual disks. On the other hand, when the data 52 is read for the third time and thereafter, it is only necessary to determine whether or not the data 52 exists by using two virtual disks. As described above, the number of virtual disks for determining the presence or absence of the data 52 is reduced, so that the data read efficiency is improved.

∙ Data read request may be made for snapshots created in the past. In the case of a data read request for a snapshot, the virtual disk that has been used as the active disk until immediately before the creation of the snapshot is the starting point for determining whether there is data.

FIG. 17 is a diagram illustrating an example of a process for reading snapshot data. In the example of FIG. 17, after the situation shown in FIG. 11 is reached, a read request 55 that requests reading of the data 52 of the offset value “xxxxx” from the snapshot (snapshot # 3) created for the third time. Is entered. In this case, the virtual disk 113 that has been used as the active disk until immediately before the third snapshot creation is the starting point for determining whether the data 52 exists. Then, it is determined whether or not data 52 is stored in each virtual disk in order from the virtual disk 113 in the order of the virtual disks. In the example of FIG. 17, it is determined that the data 52 is stored in the virtual disk 111 that is the original disk, and the data 52 is read from the virtual disk 111.

In this way, when the presence / absence of the data 52 is determined using the virtual disk 113 as a starting point, the movement number is set for each of the virtual disks 111 to 113 from the virtual disk 111 from which the data 52 has been read to the starting virtual disk 113. . Based on the set number of movements, the total number of movements corresponding to the read data is set for each virtual disk.

FIG. 18 is a diagram showing an example of setting the total number of movements when reading snapshot data. The movement number of the virtual disk 111 from which the data 52 has been read is “0”, and the movement number of the next virtual disk 112 is “1”. Therefore, in the read block management area 121a corresponding to the virtual disk 112, “1” is added to the total number of movements corresponding to the offset value “xxxxx” of the block including the data 52 and updated to “2”. The number of movements of the virtual disk 112 in the next virtual disk 113 is “2”. Therefore, in the read block management area 121b corresponding to the virtual disk 113, “2” is added to the total number of movements corresponding to the offset value “xxxxx” of the block including the data 52 and updated to “4”. At this time, there is no total number of movements exceeding the threshold “5”.

Hereinafter, it is assumed that the read request of the data 52 with the offset value “xxxxx” from the snapshot created for the third time is input again. Then, each time the data 52 is read, the total number of movements of the data 52 in the

virtual disks

112 and 113 is further updated.

FIG. 19 is a diagram illustrating an example of a second read process of snapshot data. When a third read request 56 of the data 52 from the snapshot created for the third time is input, the presence or absence of the data 52 is determined in order from the virtual disk 113 that is the starting point, as in the first time. Then, the data 52 is read from the virtual disk 111 storing the data 52. Then, in accordance with the reading of the data 52, the total number of movements of the blocks including the data 52 in the

virtual disks

112 and 113 is updated.

In the example of FIG. 19, the total number of movements of blocks including data 52 in the virtual disk 112 is updated to “3”, and the total number of movements of blocks including data 52 in the virtual disk 113 is updated to “6”. In this case, the total number of movements of the blocks including the data 52 in the virtual disk 113 exceeds the threshold “5”. Therefore, after confirming that the virtual disk 113 has sufficient free space, the block including the data 52 is copied to the virtual disk 113.

FIG. 20 is a diagram showing an example of data copy in response to reading of snapshot data. As shown in FIG. 20, the data 52 of the virtual disk 111 is copied to the virtual disk 113. The copied data is held in the virtual disk 113 as difference data. At this time, the offset value of the block including the data 52 is set in the update block list (see FIG. 6) corresponding to the virtual disk 113. Thereafter, when the third read request for the data 52 is input starting from the virtual disk 113, the data 52 is read from the virtual disk 113.

Thus, the total number of moves is also updated when snapshot data is read. That is, when the virtual disk in the logical volume 110a has a plurality of tiers due to the creation of the snapshot, the data can be read out during the operation and the snapshot data can be read out. The virtual disk that is the starting point for determining the presence / absence of data differs between when data is being read and when snapshot data is read. On the other hand, even when data is read from any virtual disk, read efficiency can be improved by copying data with high read frequency near the virtual disk that is the starting point. Therefore, in the second embodiment, the number of movements is added to the total number of movements without distinguishing which virtual disk is the read request. Thereby, by determining the necessity of copying and the copy destination based on the total number of movements, it is possible to perform appropriate copying for improving the overall reading efficiency.

[Other Embodiments]
In the second embodiment, the virtual disk used as the operation disk is excluded as the data copy destination, but the operation disk can be included in the data copy destination. In this case, for example, the access unit 140 also provides a corresponding read block management area for the operational disk, and manages the total number of movements for each data. Then, when the total number of movements on the active disk exceeds the threshold, the access unit 140 copies the read data to the active disk. For example, data may be copied to the active disk when there is not enough free space on the virtual disk whose total number of movements exceeds the threshold and data cannot be copied to the virtual disk.

In the second embodiment, the total number of data movements is calculated for each storage unit. This is also intended to improve the data reading efficiency when reading data from a created snapshot. It is. When the efficiency of reading data from the operation disk is not taken into consideration and only the efficiency of the data read process starting from the operation disk is considered, for example, the access unit 140 moves the virtual disk before the operation disk. Calculate only the total number. When the access unit 140 detects data in which the total number of movements in the virtual disk before the active disk exceeds the threshold, the access unit 140 stores the data in the virtual disk before the active disk. By doing in this way, the process of the access part 140 can be simplified and the efficiency of a process can be aimed at.

In the second embodiment, the logical volume 110a is composed of one or more virtual disks, but a physical disk or SSD (Solid State Drive) can be used instead of the virtual disk.

The above merely shows the principle of the present invention. In addition, many modifications and changes can be made by those skilled in the art, and the present invention is not limited to the precise configuration and application shown and described above, and all corresponding modifications and equivalents may be And the equivalents thereof are considered to be within the scope of the invention.

1a, 1b, 1c, 1d Storage means 2 Judgment means 3 Reading means 4 Storage means 5 Data X Information processing apparatus

Claims

A storage means is associated with each of a plurality of time zones, and the storage means stores updated data within the corresponding time zone. When a data read request is input, each storage means Determine whether the data is stored in a predetermined order,
Read the data from the storage means storing the data specified in the read request,
Storing the data in a storage means earlier in the predetermined order than the storage means;
A program that causes an information processing apparatus to execute processing.
The plurality of time zones are divided with the input time point of the snapshot creation request as a boundary,
2. The program according to claim 1, wherein the data is stored in any of storage means corresponding to a time zone before the input time point of the last snapshot creation request.
In data storage, data storage conditions are defined in advance, and the data write conditions are stored for each storage means earlier in the predetermined order than the storage means storing the data specified in the read request. 3. The program according to claim 1, wherein the data is stored in storage means that satisfies the write condition.
In the data storage, when the data is read, the predetermined number of storage units storing the data is stored for each storage unit whose predetermined order is earlier than the storage unit storing the data. A value indicating a difference in order of the data is added to a determination value indicating whether the data needs to be stored, and the data is stored in a storage unit whose determination value exceeds a predetermined threshold value. The program according to item 3.
The plurality of time zones are divided with the input time point of the snapshot creation request as a boundary,
When determining whether or not data is stored, when a read request for data held in a specific snapshot is input, the time before the input of the snapshot creation request instructing creation of the snapshot The program according to any one of claims 1 to 4, wherein the storage unit corresponding to the band sequentially determines whether or not the data is stored in the predetermined order. .
Information processing device
A storage means is associated with each of a plurality of time zones, and the storage means stores updated data within the corresponding time zone. When a data read request is input, each storage means Determine whether the data is stored in a predetermined order,
Read the data from the storage means storing the data specified in the read request,
Storing the data in a storage means earlier in the predetermined order than the storage means;
A data management method characterized by the above.
A storage means is associated with each of a plurality of time zones, and the storage means stores updated data within the corresponding time zone. When a data read request is input, each storage means Determining means for determining whether or not the data is stored in a predetermined order;
Reading means for reading the data from the storage means storing the data specified in the read request;
Storage means for storing the data in the storage means earlier in the predetermined order than the storage means;
An information processing apparatus.