JP3882461B2 - Storage device system and backup acquisition method thereof - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a storage device system in a computer system, and more particularly to a storage device system capable of acquiring a data backup without using a host computer.
[0002]
[Prior art]
In general, in a computer system, data stored in a storage device is backed up in order to prevent important data from being completely lost when a failure occurs in the storage device. In order to guarantee the consistency of the data to be backed up, the data write process to the storage device usually has to be stopped while the backup is taken.
[0003]
For example, Japanese Patent Laid-Open No. 7-210439 discloses a storage device that can solve such a problem and can perform backup processing without interposing a host computer.
[0004]
The storage device disclosed in Japanese Patent Application Laid-Open No. 7-210439 includes (1) means for managing an area to be backed up separately as a backed up area and an unbackup area, and (2) writing to the unbackup area. Means for reading the data recorded in the area to be written and saving it in the buffer when the request is received, and then receiving the write data and recording it in the storage medium; (3) to be backed up Means for identifying whether the data to be transferred is in a storage medium or in a buffer when transferring the data to another storage device, and (4) the data saved in the buffer is appropriate Means for positively transferring the saved data to another storage device at the backup destination when exceeding a predetermined amount, and (5) another storage device at the backup destination When a sequential access type, comprises means for transferring the data, together with position information indicating where the was recorded in the data storage medium to another storage device.
[0005]
In this storage device, in order to prevent the buffer from being filled by the above means (4), when the storage device for storing the backup data is in the sequential access format, there is a problem that it is not recorded in the correct order. With respect to this, it is possible to restore based on the position information without storing the data in the sequential access type storage device in the correct order by means (5).
[0006]
[Problems to be solved by the invention]
However, in the technique disclosed in Japanese Patent Laid-Open No. 7-210439 described above, when the storage device for storing backup data is in the sequential access format, processing overhead for creating data location information is required. Further, it is necessary to transfer and store the position information in addition to the data, and accordingly, there is a problem that a storage capacity for the backup data is required.
[0007]
Furthermore, in the technique disclosed in Japanese Patent Laid-Open No. 7-210439, when a write request occurs during backup, the data before being updated by the write request is saved in a buffer, so that the data is transferred to the backup destination storage device In this case, it is necessary to manage whether the data to be transferred is in the storage medium or in the buffer. There is also a problem that there are many management items such as which area of the buffer is open to save the data before update, and which area of the data is saved.
[0008]
An object of the present invention is to provide a storage device that stores backup data in the correct order and eliminates the need to record data location information even when the storage device for storing backup data is a sequential access type storage device. It is in.
[0009]
Another object of the present invention is to provide a storage device that simplifies management of saved data.
[0010]
[Means for Solving the Problems]
In order to achieve the above object, the present invention is a function for acquiring a backup by transferring data stored in an area indicated by an instruction in a volume to another storage device in response to an instruction from a host. In a storage device system that has data storage, the specified area is copied to an area that has already been copied to another storage device, and the data is not copied to another storage device, and data is saved to a volume for saving data. Area management means for managing separately, and other storage areas that have not been copied to other storage devices and that have not been backed up to a data saving volume When a write request is received for an area that has not been copied to the device and data has not been saved to the save volume, the write request is retained in the target area of the write request. Data stored in the data save volume, then write request execution means for receiving the write data, whether the data to be transferred to another storage device is in an area in the volume or in the data save volume A copy source management unit that manages the data, and a copy unit that transfers the location data obtained by the copy source management unit to the storage device designated by the instruction.
[0011]
Further, according to another aspect of the present invention, a user volume storing user data used by a computer, and for saving data when a backup is obtained by copying the user volume data to another storage device There is provided a backup method in a storage system having a work volume to be used and a control device for controlling access to the user volume and the work volume. In this backup method, in response to a data copy request from the user volume to another storage device, the user data held in the area specified by the copy request from the user volume is copied to the other storage device. . During a copy process to another storage device, when there is a data write request to a part of the specified area, the user data held in that part of the area is transferred to the work volume. After that, data is written to some areas. In the copy process, it is determined whether the user data to be copied to another storage device is held in an area in the user volume or an area in the work volume, and depending on the result of this determination The user data to be copied is read and transferred to another storage device.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a block diagram showing a configuration example of a computer system including a storage device to which the present invention is applied.
[0013]
The computer system of this embodiment includes a disk subsystem 118 that stores user data, and a host that issues read requests, write requests, copy requests, and the like for the user data stored therein to the disk subsystem 118. A computer 116 and a magnetic tape device 117 that stores backup data of data stored in the disk subsystem 118 are included.
[0014]
The disk subsystem 118 includes a host interface 102 serving as an interface with the host computer 116, a magnetic tape device interface 104 serving as an interface with the magnetic tape device 117, a control processor 108 for controlling various processes performed in the disk subsystem 118, A control memory 109 for storing control information used by the control processor 108 and a volume access unit 114 for accessing a volume as a storage device for storing data are provided.
[0015]
The control processor 108 includes a volume attribute change process 101, a write process 103, and a copy process 105 as processing functions related to the backup process in this embodiment. Specifically, these processing functions are realized by a program executed by the control processor 108. In addition, the control processor 108 has various functions for controlling the disk subsystem, but since these are not directly related to the present invention, illustration and description are omitted here. Of these processing functions, the volume attribute change processing 101 is not used in the first embodiment. The volume attribute change process 101 will be described in a second embodiment described later.
[0016]
The control memory 109 holds volume management information 106 and copy status management information 107 as control information used in the backup processing in this embodiment.
[0017]
The disk subsystem 118 includes user volumes 110 and 111 for storing user data, and work volumes 112 and 113 for storing saved data during backup processing, as volumes that are storage devices for storing data. Each volume is given a volume number as an identifier for identifying each volume.
[0018]
In this embodiment and a second embodiment described later, a volume refers to a logical storage device that is formed using a physical disk device and is recognized as one storage device by a host computer. One physical disk device may be directly used as one volume, or a plurality of volumes may be formed on one disk device. When the disk subsystem 118 is configured as a so-called RAID, a plurality of volumes are formed on a plurality of physical disk devices configuring the RAID.
[0019]
The storage area of each volume is composed of a plurality of blocks. Each block is given a block number as information for identifying each block. The volume includes blocks whose block numbers take values of 0, 1, 2,..., N (N is an integer value of 0 or more and may be different for each volume).
[0020]
FIG. 2 is a conceptual diagram showing the data structure of the copy status management information 107.
[0021]
The copy status management information 107 includes information related to the copy operation. The copy status management information 107 is referred to and updated in a write request issued by the host computer 116, a write process 103 executed in response to the copy request, and a copy process 105.
[0022]
As shown in the figure, the copy status management information 107 includes a busy flag 300, a copy source volume number 303, and block information 304. The in-use flag 300 indicates whether or not the copy status management information 107 is in use, and has a value indicating either “unused” or “in use”. The copy source volume number 303 stores the volume number of the volume that stores the data to be copied by the backup process. The block information 304 is information indicating the block status of the volume indicated by the copy source volume number 303.
[0023]
The block information 304 includes a block number 301 and a copy state 302 as information indicating the block state. The copy status 302 has a value indicating one of “copy not required”, “copy required & not saved”, “copy required & saved”, and “copied”.
[0024]
“Copy unnecessary” indicates that the block stores data that does not need to be copied. “Copy required & not saved” indicates that the block stores data that currently needs to be copied. “Copy Required & Saved” indicates that the block storing data that needs to be copied at the time of receiving the copy request, but that data has been saved to the work volume.
[0025]
FIG. 3 is a conceptual diagram showing the data structure of the volume management information 106.
[0026]
The volume management information 106 is management information indicating the status of the volume, and has a copy operation presence / absence flag 401 for each volume number 400. The copy operation presence / absence flag 401 takes a value indicating either “copying” or “normal”. “Copying” indicates that the volume is a copy process target. “Normal” indicates that the volume is not a copy process target.
[0027]
FIG. 4 is a flowchart showing the flow of the write process 103 executed in response to a write request from the host computer in this embodiment.
[0028]
The host computer 116 issues a write request to the disk subsystem 118 when attempting to write or update user data. The write request includes the volume number of the volume in which the write data is written, the block number of the block in which the write data is written, and the write data.
[0029]
When receiving the write request, the control processor 108 starts processing by the write processing 103.
[0030]
First, the control processor 118 refers to the volume management information 106 and determines whether copy processing described later is being executed on the data of the volume that is the target of the write request. Specifically, the status of the copy operation presence / absence flag 401 is checked. If the copy operation presence / absence flag 401 indicates “copying”, the process proceeds to step 501; if it indicates “normal”, the process proceeds to step 504 (step 500).
[0031]
In step 501, the control processor 118 refers to the copy status management information 107 to check the status of the write request, and determines whether data saving is necessary. Specifically, the copy status management information 107 that stores the same value as the volume number of the volume that is the target of the write request in the copy source volume number 303 is selected. Next, the control processor 118 checks the copy status 302 to check whether the block that is the target of the write request is “copy required & not saved”. If the copy status 302 indicates “copy required & not saved”, the process proceeds to step 502 and data is saved. If the copy status 302 is a value indicating anything other than “copy required & not saved”, it is not necessary to save data, and the process proceeds to step 504.
[0032]
In step 502, first, of the block that is the target of the write request, the data of the block that is indicated as “copy required & not saved” by the copy status 302 is read from the user volume 110 by the volume access unit 114. Next, the control processor 118 records the read data in the work volume 112. At this time, the control processor 118 writes the data to the block of the work volume 112 having the same block number as the block in which the data read in the user volume 110 was stored.
[0033]
In step 503, the control processor 118 updates the copy status 302 of the block whose data was saved in step 502 to a value indicating “copy required & saved”.
[0034]
In step 504, the write data received from the host computer 116 by the write request is recorded in the designated block of the user volume 110, and the process is terminated.
[0035]
FIG. 5 is a flowchart showing the flow of the copy process 105 executed in response to a copy request from the host computer in this embodiment.
[0036]
The copy request issued by the host computer 116 has the volume number of the volume storing the data to be copied and the block number of the block storing the data to be copied.
[0037]
When receiving a copy request from the host computer 116, the control processor 118 starts processing by the copy processing 105.
[0038]
In the copy process 105, the control processor 118 selects an entry from the volume management information 106 in which the volume number is 400 and the same volume number as the received volume number is set as a copy request parameter. Then, a value indicating “copying” is set in the copy operation presence / absence flag 401 of the entry (step 600).
[0039]
The control processor 118 selects copy status management information in which a value indicating “unused” is set in the in-use flag 300 from the copy status management information 107, and sets the in-use flag of the selected copy status management information 107. Set a value indicating "in use". Next, a volume number as a copy request parameter is set in the copy source volume number 303. Next, the control processor 118 sets a value indicating “copy required & not saved” in the copy status 302 corresponding to the same block number 301 as the block number specified by the parameter of the copy request, and sets it to another block number 301. A value indicating “copy unnecessary” is set in the corresponding copy status 302 (step 601).
[0040]
Thereafter, the variable i for designating a block is set to 0 and the variable i is initialized (step 602).
[0041]
The control processor 118 determines the next process to be executed according to the value of the copy status 302 corresponding to the same block number 301 as the value of the variable i in the copy status management information 107 selected at step 601. That is, if the copy status 302 indicates “copy required & not saved”, the process proceeds to step 604, and if “copy required & saved” is indicated, the process proceeds to step 605. If the copy status 302 indicates “copy not required” or “copied”, the process proceeds to step 607 (step 603).
[0042]
In step 604, the control processor 118 reads data from the block of the user volume 110 having the same block number as the variable i using the volume access unit 114. The control processor 118 transfers the data read from the user volume 110 to the magnetic tape device 117 via the magnetic tape device interface 104 and records it on the magnetic tape.
[0043]
In step 605, data is read from the block of the work volume 112 having a block number that matches the variable i using the volume access unit 114. The control processor 118 transfers this data to the magnetic tape device 117 via the magnetic tape device interface 104 and records it on the magnetic tape.
[0044]
After the data transfer from the user volume 110 or the work volume 112 to the magnetic tape device 117, the control processor 118 indicates that the copying of the data of the block has been completed. A value indicating “copied” is set in the copy status 302 corresponding to the same block number 301 (step 606).
[0045]
In step 607, it is determined whether processing has been executed for all blocks of the user volume 110. If the process is completed for all blocks, the process proceeds to step 609. If there are still remaining blocks, the control processor 118 increments the value of the variable i by 1 in step 608 and then returns to the processing in step 603 to perform the same processing for the remaining blocks.
[0046]
In step 609, a value indicating “unused” is set in the in-use flag 300 of the copy status management information 107 selected in step 601. Finally, the control processor 118 sets a value indicating “normal” in the copy operation presence / absence flag 401 of the volume management information 106 and ends the copy process 105.
[0047]
In the embodiment described here, copy processing is performed in response to a copy request issued from the host computer 116. In addition, the disk subsystem 118 may include the operation terminal 100 serving as an interface with the operation terminal 115 and issue a copy request from the operation terminal 115. In this case, the control processor 108 receives a copy request from the operation terminal 115 via the operation terminal interface 100 and starts the copy process 105.
[0048]
As described above, by using a volume as an area for saving data, the cost can be reduced as compared with a case where a semiconductor memory such as a buffer is used as a storage medium for saving data. Further, since the data is backed up in the order specified by the copy request, the backup data can be stored in a sequential access type storage medium such as a magnetic tape without using position information. As a result, it is possible to reduce the overhead of backup processing, the amount of transferred data, and the amount of stored data.
[0049]
Next, another embodiment of the present invention will be described.
[0050]
The computer system according to the second embodiment described below basically has the same configuration as that of the first embodiment. In this embodiment, a process using the volume attribute change process 101 not used in the first embodiment will be described. By using the volume attribute change processing 101, the contents of the control information and the processing to be performed are slightly different from those in the first embodiment.
[0051]
FIG. 6 is a conceptual diagram showing the data structure of copy status management information used in this embodiment. As shown in the figure, in the copy status management information of this embodiment, a work volume number 700 is added to the in-use flag 300, the copy source volume number 303, and the block information 304.
[0052]
The work volume number 700 is a volume number used as a work volume for saving data during execution of a copy request from the host computer 116 or the operation terminal 115. In this embodiment, the volume having the volume number held in the work volume number 700 is used as the work volume.
[0053]
FIG. 7 is a conceptual diagram showing the data structure of the volume management information 106 in this embodiment. As shown in the figure, the volume management information 106 of this embodiment has an attribute flag 800 added to the volume management information used in the first embodiment.
[0054]
A value indicating either “user” or “work” is set in the attribute flag 800 as a flag indicating the attribute of the volume specified by the volume number 400. When the value set in the attribute flag 800 indicates “user”, the volume is used as a user volume for storing user data. On the other hand, when the attribute flag 800 is a value indicating “work”, the volume is used as a work volume for storing saved data.
[0055]
The host computer 116 or the operation terminal 115 can change a volume attribute (attribute such as user volume or work volume) by issuing a volume attribute change command to the disk subsystem 118. The volume attribute change command includes “volume number” and “attribute” as parameters. The parameter “volume number” is a volume number indicating a volume whose attribute is to be changed. The parameter “attribute” indicates a new attribute of the volume designated by the volume number.
[0056]
FIG. 8 is a flowchart of the volume attribute change processing 101 executed by the control processor 108 in response to the volume attribute change command.
[0057]
In the volume attribute change processing 101, the control processor 108 first refers to the volume management information 106, and checks the volume copy operation presence / absence flag 401 corresponding to the parameter “volume number” of the received volume attribute change command (step). 1000).
[0058]
When the copy operation presence / absence flag 401 is “copying”, the control processor 108 notifies the requesting host computer 116 or the operation terminal 115 that the attribute cannot be changed, and ends the processing (step 1002). ).
[0059]
In step 1000, if the copy operation presence / absence flag 401 is "normal", the control processor 108 changes the attribute flag 800 to the attribute specified by the parameter and ends the process (step 1001).
[0060]
FIG. 9 is a flowchart of the write process in this embodiment. In the write process in the present embodiment, as shown in the drawing, the process in step 502 in the write process in the first embodiment is replaced with the processes in step 1100 and step 1101.
[0061]
When the data needs to be saved (the control processor is executing a copy request for the volume that is the target of the write request, and the copy status 302 of the target block of the write request is “copy required & not saved”. In step 1100, the copy status management information 107 in which the in-use flag 300 is “in use” and the volume number of the volume targeted for the write request is set in the copy source volume number 303 is acquired. Furthermore, the control processor identifies a volume to be used as a work volume for saving data based on the volume number set in the work volume number 700 of the corresponding copy status management information 107.
[0062]
In step 1101, as in step 502 of the first embodiment, data is read from the copy source volume, and the data is written to the work volume identified in step 1100.
[0063]
The processes with the same step numbers as in FIG. 4 other than steps 1100 and 1101 are the same as the processes of the corresponding steps described in the first embodiment, and the description thereof is omitted here.
[0064]
FIG. 10 is a flowchart of the copy process in this embodiment.
[0065]
The copy process according to the present embodiment is different from the copy process according to the first embodiment in the following points. Instead of step 601 in the first embodiment, processes of steps 1200 and 1201 are performed. The process of step 1202 is performed instead of the process of step 605 in the first embodiment. And the process of step 1203 is added at the end of the process.
[0066]
In step 1200, the control processor 108 refers to the volume management information 106 and selects a volume whose copy operation presence / absence flag 401 is “normal” and whose attribute flag 800 is “work”, that is, an unused work volume. Then, a value indicating “copying” is set in the copy operation presence / absence flag 401 corresponding to the selected volume.
[0067]
Next, in step 1201, the control processor 108 selects the copy status management information 107 to be used and sets each piece of information in the same manner as in step 601 in the first embodiment. At this time, in this embodiment, the volume number of the volume selected in step 1200 is set in the work volume number 700 of the copy status management information 107.
[0068]
In step 1202, the control processor 108 reads data stored in the block having the same block number as the variable i from the volume selected in step 1200. The control processor 108 then transfers the data read via the magnetic tape device interface 104 to the magnetic tape device 117 and records the data on the magnetic tape.
[0069]
In step 1203, “normal” is set in the copy operation presence / absence flag 401 corresponding to the volume selected in step 1200 of the volume management information 106, and the in-use state is released.
[0070]
For processes other than those described here, the same process as the copy process in the first embodiment is performed. Therefore, the description is omitted here.
[0071]
According to the present embodiment, a command for changing the attribute of a volume is supported, and a user volume for storing user data and a work volume for storing saved data can be switched and used. Therefore, the number of work volumes can be changed in accordance with the required multiplicity of copy requests, and volume resources can be used effectively.
[0072]
Also, when a copy request is received, the work volume used for executing the copy request is selected and used from the unused work volume group by the disk subsystem. When the copy processing is completed, the used work volume is returned to an unused state. In this way, the work volume can be effectively utilized by reusing one work volume.
[0073]
Furthermore, since the program that issues a copy request on the host computer or operation terminal side does not need to manage the work volume, the program becomes simple.
[0074]
In the first and second embodiments described above, the save data is stored in the block having the same block number as the user volume in the work volume. However, the save data can also be stored in order from the head of the work volume. In this case, the copy status management information 107 may have information on which block has been saved in addition to the information on whether or not it has been saved. In the copy processing, the location where the saved data is stored is checked from this information, the data is read from the corresponding block, and the data is transferred to the magnetic tape device. By doing so, it is sufficient to secure an area for storing saved data by the number of blocks indicated by the copy request from the host computer or operation terminal, and multiple copy requests are executed with a smaller work volume. can do.
[0075]
When the disk subsystem receives a request for access to a volume from the host computer, the volume management information is referenced to check whether the volume is a user volume. If the volume is a user volume, an access request is issued. When the volume is a work volume, the host computer can be notified that the volume cannot receive an access request. By providing such a check mechanism, the disk subsystem can prohibit access to an unauthorized volume.
[0076]
According to the embodiment described above, for the block indicated by the copy request, the disk sub that backs up the data at the time of receiving the copy request to the magnetic tape device or the like without receiving the host while receiving the write request from the host. In the system, the following effects can be achieved.
[0077]
First, by using a volume as an area for saving data, the cost can be reduced compared to a semiconductor memory such as a buffer.
[0078]
Second, backup data can be stored in a sequential access format storage medium such as magnetic tape in the order indicated by the copy request, so location information indicating the location where the backed-up data was stored is added. Therefore, it is possible to reduce the overhead of creating location information, reduce the amount of transferred data, and reduce the amount of stored data.
[0079]
【The invention's effect】
According to the present invention, it is possible to acquire a copy of user data at a certain point in time while continuously receiving data write requests from the host computer.
[Brief description of the drawings]
FIG. 1 is a simplified block diagram showing a configuration example of a computer system including a storage device to which the present invention is applied.
FIG. 2 is a conceptual diagram showing a data structure of copy status management information in the first embodiment.
FIG. 3 is a conceptual diagram showing a data structure of volume management information in the first embodiment.
FIG. 4 is a flowchart showing a flow of write processing in the first embodiment.
FIG. 5 is a flowchart showing a flow of copy processing in the first embodiment.
FIG. 6 is a conceptual diagram showing a data structure of copy status management information in the second embodiment.
FIG. 7 is a conceptual diagram showing a data structure of volume management information in the second embodiment.
FIG. 8 is a flowchart showing the flow of a volume attribute change process.
FIG. 9 is a flowchart showing a flow of write processing in the second embodiment.
FIG. 10 is a flowchart showing a flow of copy processing in the second embodiment.
[Explanation of symbols]
101 ... Volume attribute change processing, 103 ... Write processing, 105 ... Copy processing, 106 ... Volume management information, 107 ... Copy status management information, 108 ... Control processor, 109 ... Control memory, 110, 111... User volume, 112, 113... Work volume, 116... Host computer, 117.

Claims (4)

In a storage system having a function of acquiring a backup by transferring data stored in an area indicated by the instruction in the volume to another storage device in response to an instruction from the host.
A first interface connected to the host and receiving instructions from the host;
A second interface connected to the other storage device and transferring data;
Work volume management means that has a plurality of data saving volumes and manages the usage status of the plurality of data saving volumes, and unused data managed by the work volume management means in response to a copy request A data evacuation volume allocation means for selecting an evacuation volume and allocating it for data evacuation;
The designated area includes an area where data copy to the other storage device has been completed, and an area where data has not been copied to the other storage device and data has been saved to the data saving volume. An area management means for managing the data separately from an area where data has not been copied to the other storage device and data has not been saved to the data saving volume;
When a write request is received for an area that has not been copied to the other storage device and data has not been saved to the save volume, the data held in the write request target area Write request execution means for receiving write data after storing the data in the data saving volume;
Copy source management means for managing whether the data to be transferred to the other storage device is in an area in the volume or in the data saving volume;
Copy means for transferring the location data obtained by the copy source management means to the other storage device designated by the instruction via the second interface ;
A data evacuation volume releasing means for returning the data evacuation volume used for data evacuation to an unused state in response to execution of the copy request;
A storage system comprising:   The storage system according to claim 1, wherein the copy unit transfers data to the storage device in an order specified by the instruction.   2. The storage system according to claim 1, further comprising volume switching means for receiving a command for switching between a user volume storing user data and a work volume storing saved data.   When the write request execution means saves data to the data saving volume, the write request execution means stores the data in an area in the data saving volume that has the same address as the area in the volume in which the data to be saved is stored. The storage system according to claim 1, wherein the storage system is evacuated.

Images