This application relates to and claims the benefit of priority from Japanese Patent Application number 2007-98052, filed on Apr. 4, 2007, the entire disclosure of which is incorporated herein by reference.
The present invention relates to a computer system comprising a computer which executes an application program.
Technology called CDP (Continuous Data Protection) is known. CDP-related technology, for example, is disclosed in Japanese Patent Laid-open No. 2005-18738. According to this Japanese Patent Laid-open No. 2005-18738, a journal is created and stored each time a storage system receives a write request from a host computer and writes data to a logical volume, and when a restore is performed, a journal is used to restore data from an arbitrary time to a logical volume.
For example, there is a computer system, which comprises a storage system comprising a plurality of logical volumes; a computer (hereinafter, job host) for executing one or more job application programs (hereinafter, job AP) using the plurality of logical volumes; and a computer (hereinafter, storage management host) for managing the plurality of logical volumes. In an environment in which this computer system is operated, there may be an administrator (hereinafter, storage administrator), who uses the storage management host to manage the plurality of logical volumes, and an administrator (hereinafter, AP administrator), who manages a job AP, which is being executed by the job host.
In this computer system, the storage management host can send to a storage system a restore indication, which specifies the storage administrator's arbitrary RPO (Recovery Point Objective) time, in accordance with an operation by the storage administrator. In this case, the storage system, in response to this restore indication, can restore data of the specified RPO time to a logical volume.
Here, for example, when a failure occurs in the storage system while a job is being performed, and the storage administrator restores data in the storage system up to an arbitrary RPO time, the AP administrator does not know what data was restored. Further, the AP administrator also does not know if the restored data affects an AP, which he is managing. When these points need to be specified, the AP administrator must query the storage administrator.
Further, for example, the RPO time does not necessarily have to be an arbitrary time of the storage administrator, but can also be an arbitrary time decided by the AP administrator. However, the AP administrator is not aware of the utilization status of a logical volume by a job AP other than the job AP that he himself is managing. For example, there are times when a logical volume being used by an AP administrator-managed AP is also being used by a job AP other than the job AP that the AP administrator is managing himself. Thus, it is difficult for the AP administrator to decide an RPO time.
The above-mentioned problems can also exist when an AP is not a job AP.
Therefore, an object of the present invention is to support the administrator of an application program that utilizes restored data.
Other objects of the present invention should become clear from the following description.
A first information resource denoting which logical volume is allocated to which application program is prepared by a management computer. The management computer either regularly or irregularly acquires from the storage system information as to which logical volumes were updated at what times, registers the acquired information in a second information resource, references the first and second information resources, acquires update management information, which is information denoting which logical volume is updated at what time, and the application program to which this logical volume is allocated, and sends this update management information to a host computer. The host computer, based on the update management information from the management computer, displays which logical volume is updated at what time, and which application program is allocated to this logical volume.
In one embodiment, each of one or more application programs executed by a host computer is constituted so as to be able to issue a write request specifying a logical volume allocated from among a plurality of logical volumes in a storage system (a storage system having a CDP (Continuous Data Protection) function for restoring data at an RPO (Recovery Point Objective) time inside a specified logical volume). A management computer of the storage system comprises a first information resource for denoting which logical volume is allocated to which application program; a volume update manager, which either regularly or irregularly acquires from the above-mentioned storage system what unit of time updating took place in which logical volume, and registers the acquired information in a second information resource; an update management information acquisition unit, which references the first and second information resources, and acquires update management information, which is information denoting what time (for example, what unit of time) updating took place in which logical volume, and to which application program this logical volume is allocated; and an update management information sending unit for sending acquired update management information to the host computer having the application program denoted by this update management information. The host computer comprises a receiver for receiving update management information from the management computer; and a host controller for displaying, based on received update management information, what time (for example, what unit of time) updating took place in which logical volume, and which application program is allocated to this logical volume. For example, application program identification information (for example, an application name and/or another type of identification information), and logical volume identification information (for example, a LUN (Logical Unit Number)) are registered in the first information resource. A logical volume identifier, and information denoting whether or not updating took place at respective units of time are registered in the second information resource.
In one embodiment, the management computer can further comprise a failure detector for detecting a failure that occurs in a storage system. In this case, the above-mentioned update management information, which is sent, comprises failure time. The host controller can display the failure time comprised in the update management information. The management computer can further comprise a restore controller for sending to the storage system a restore request, which comprises identification information for the logical volume of the storage system and RPO time. In this case, the sent update management information comprises the RPO time. The host controller can display the RPO time comprised in the update management information.
In one embodiment, there are a plurality of host computers. The update management information acquisition unit can specify an affected volume, which is a logical volume for which updating took place at a time (for example, a unit of time) comprising a failure time, by referencing the first information resource. Further, the update management information acquisition unit can specify an affected application, which is the application program to which this logical volume is allocated, by referencing the second information resource. Further, the update management information acquisition unit can acquire, as update management information, information denoting what time (for example, what unit of time) updating took place in which logical volume, and which application program is the affected application to which this affected volume is allocated. The update management information acquisition unit can send update management information to the affected application host computer of a plurality of host computers.
In one embodiment, each of one or more application programs can register the start time and end time of an operation related to the generation of a write request in a log information resource. The host controller can display, based on the log information resource, which application program was running during what time period, together with what time (for example, what unit of time) updating took place in which logical volume, and which application program is allocated to this logical volume.
In one embodiment, there are a plurality of host computers. A host controller can conceal updating at a time (for example, a unit of time) other than the period of time during which an application program was running even when updating took place in a logical volume to which this application program is allocated.
In one embodiment, a host controller is constituted so as to receive the specification of an RPO time, and to notify the above-mentioned management computer of the specified RPO time. This host controller can refuse a specification, which is a unit of time other than a period of time during with an application program was running, and which makes the RPO time a time in the unit of time during which updating took place in a logical volume to which this application program is allocated.
In one embodiment, a host controller can receive an RPO time specification via a GUI (Graphical User Interface). A first display of what unit of time updating took place in which logical volume, and which application program is allocated to this logical volume, and a second display of which application program was running during what time period are lined up side-by-side in the GUI.
In one embodiment, each of a first display and a second display is a graph or a table. A location equivalent to a certain time in the first display, and a location equivalent to this certain time in the second display are lined up in the GUI perpendicularly or horizontally.
In one embodiment, a host controller can receive the input of application program identification information to be managed, register the inputted application program identification information in a prescribed information resource, and notify the management computer. The management computer can further comprise an application identification information receiver for receiving application program identification information; and a first information resource updating unit for registering in the first information resource application program identification information and identification information for the logical volume allocated to this application program.
In one embodiment, a host controller can compute and display a recommended RPO time based on update management information from the management computer, and the above-mentioned log information resource.
In one embodiment, the management computer can further comprise a third information resource for denoting which logical volume belongs to which volume group. The update management information comprises information denoting which logical volume belongs to which volume group. The host controller can also display which logical volume belongs to which volume group.
Two or more of the above-described plurality of embodiments can be arbitrarily combined.
The above-mentioned information resource is electronic information stored in a storage resource (for example, a memory or other type of storage device), and, more specifically, for example, is a file or table. Further, the above-mentioned respective units (for example, the volume update manager, update management information acquisition unit, update information transmitter, receiver, and host controller) can be constructed using hardware, a computer program, or a combination thereof (For example, a portion can be realized using a computer program, and the remainder can be realized via hardware.). A computer program is read into a prescribed processor and executed. Further, when a computer program is read into a processor to carry out information processing, a storage area, which resides in a memory or some other such hardware resource, can be used as deemed appropriate. Further, a computer program can be installed into a computer from a CD-ROM or other such storage medium, or it can be downloaded to a computer via a communication network.
A number of embodiments of the present invention will be explained in detail hereinbelow by referring to the figures.
<First Embodiment>
<<System Configuration>>
Two (can also be either one, or three or more) job hosts 101, one (can also be more than one) storage management host 103, and two (can also be either one, or three or more) storage systems 105 are connected to a first communication network (for example, a LAN (Local Area Network)) 109. The respective job hosts 101, storage management host 103, and respective storage systems 105 can exchange various information via the first communication network 109. A storage administrator 113 operates the storage management host 103.
Two job hosts 101 and two storage systems 105 are connected to a second communication network (for example, a SAN (Storage Area Network)) 107. The respective job hosts 101 can access (write or read data) a specified logical volume by issuing an access request, which specifies an access destination logical volume in the respective storage systems 105 via the second communication network 107.
A job host 101 is a computer for executing no less than one job application program (hereinafter, will be described simply as “AP”). A job host 101 comprises a first I/F 151, which is a communication interface device for the first communication network 109; a second I/F 153, which is a communication interface device for the second communication network 107; a CPU 123; and a storage resource (for example, a memory and/or some other type of storage device) 121. In the storage resource 121 are stored no less than one AP 119 for executing a job; an AP management program 117 for managing the no less than one AP 119; and an AP registration table 115 in which information related to the no less than one AP 119 is recorded. In the example shown in the figure, three APs 119 (AP 1, AP 2, AP 3), an AP management program 1, and an AP registration table 1 are stored in storage resource 1 of job host 1, and two job APs 119 (AP 4, AP 5), an AP management program 2, and an AP registration table 2 are stored in storage resource 2 of job host 2. Hereinafter, when a computer program is the subject of a sentence, it is supposed that processing is actually carried out by the CPU, which executes this computer program. The AP management program 117 and the AP registration table 115 will be explained in detail further below.
The storage management host 103 is a computer for managing respective logical volumes 181 inside the respective storage systems 105. The storage management host 103 comprises a first I/F 201, which is a communication interface device for the first communication network 109; a CPU 143; and a storage resource (for example, a memory and/or some other type of storage device) 141. A data update history table 137, a CDP repository 135, which is a computer program for updating the data update history table 137, an AP management table 133, which is a table for managing an AP 119, a DVOL management table 161, which is a table for managing a data volume 181, and a storage management program 131 are stored in the storage resource 141. The respective elements stored in the storage resource 141 will be explained in detail further below.
A storage system 105 comprises a controller 183; and a storage device unit 185.
The storage device unit 185 is constituted from a plurality of storage devices (for example, hard disk drives and/or flash memory drives). A plurality of logical volumes are formed based on the storage space of the plurality storage devices. The plurality of logical volumes, for example, comprise a data volume (DVOL) 181, which is a logical volume that is used by an AP 119; and a journal volume (JVOL) 180, which is a logical volume in which a journal, which is created each time data is written to a DVOL 181, is stored.
The controller 183, for example, is either one or a plurality of circuit boards, and comprises a first I/F 207, which is a communication interface device for the first communication network; a second I/F 209, which is a communication interface device for the second communication network; a third I/F 205, which is a communication interface device for the respective storage devices inside the storage device unit 185; a memory 189; and a CPU 187. A disk control program 191 and a restore control program 193 are stored in the memory 189. The disk control program 191, upon receiving a write request from a job host 101, writes data in accordance with this write request to the DVOL 181 specified in this write request, and at this time, creates a journal comprising this data, and writes the created journal to the JVOL 180. The restore control program 193, upon receiving a restore indication from the storage management host 103, responds to this restore indication by using a journal inside the JVOL 180 to restore the data for the RPO time specified by this restore indication.
The JVOL 180 shown in the figure takes JVOL 1 as an example. The JVOL 180 is configured by a management area 401 and a data area 403. A journal for data written to a DVOL 181 (updated data) is configured from this data (updated data) and management information related to the management of this data, and the management information is written to the management area 401, and the data is written to the data area 403. The management information, for example, comprises a VOL identifier (for example, a LUN), which shows the DVOL 181 into which the journal for the data is written; an update location (for example, logical block address) showing where in the DVOL 181 the data has been written; an update time (for example, year, month, day, hour, minute and second); a sequence number (SEQ#) showing the number of the storage system 107 in which the journal was created; and a pointer showing where in the data area 403 the data resides. The update time can be the time (time stamp) comprised in the write request from a job host 101, or the time at which the journal was created.
The AP registration table 115 shown in the figure takes AP registration table 1 as an example. The AP registration table 115 is referenced and updated by the AP management program 117. For example, a management AP name and log file pointer for each AP 119 are recorded in the AP registration table 115. The management AP name is the name of an AP targeted for management by the AP administrator 111. The log file pointer shows the pointer for a log file updated by an AP. When the AP 1 log file is to be referenced, the AP management program 117, for example, can access and reference the AP 1 log file from the log file pointer corresponding to the AP 1 name (management AP name).
The AP management table 133 is referenced and updated by the storage management program 131. The AP management table 133 is a table for managing which DVOL 181 is allocated to which AP 119 residing in which job host 101. More specifically, for each DVOL 181, for example, a VOL identifier (for example, “Volume 1”), and a AP name (for example, “AP 1”) and host name (for example, “host 1”) to which this DVOL 181 is allocated are recorded in the AP management table 133.
The DVOL management table 161 is a table for managing which DVOL 181 resides in which storage system 105. More specifically, for example, for each DVOL 181, a VOL identifier (for example, “volume 1”) and a storage identifier (for example, “storage 1”), which is the identifier of the storage system 105 having this DVOL 181, is recorded in the DVOL management table 161.
The data update history table 137 is a table for managing what time there was an update (in other words, the occurrence of a difference) in which DVOL 181. More specifically, for example, the presence or absence of a difference is denoted for each DVOL 181 in the data update history table 137 at every unit of time (for example, every one second). In this embodiment, when there is a difference, a value signifying a difference (for convenience sake, called a “difference mark” hereinafter) is recorded, and when there is not a difference, nothing is recorded. However, the present invention is not limited to this, and another method can be used to denote when a difference occurred in which DVOL 181. For example, a method in which only the time when a difference occurred or the time when a difference did not occur for each DVOL 181 can be used. Further, in
The data update history table 137 can be manipulated by the CDP repository 135, and the storage management program 131 can utilize this CDP repository 135 to reference and update the data update history table 137.
<<Premise>>
An explanation of the premise of the computer system related to the first embodiment will be given hereinbelow.
The AP management program 1 responds to an indication from the AP administrator 1, receives the registration of an AP targeted for management (for example, displays the registration of the AP on the AP registration screen 501 shown in
Thereafter, the AP administrator 1 sends a DVOL allocation request comprising the inputted AP name “AP 1” to the storage administrator 113 (For example, the AP administrator 1 notifies the storage administrator 113 via electronic mail mentioning the DVOL allocation request and the inputted AP name “AP 1”.). The storage administrator 113, who receives the request, uses the storage management program 131 to register the AP name received together with the request, the host name of the request-source job host 101, and the VOL identifier of the DVOL 181 to be allocated to this AP name in the AP management table 133. If a DVOL 181, which is allocated to at least one AP 119, is allocated, the VOL identifier of this DVOL 181, the AP name received together with the request, and the host name of the request-source job host 101 are made correspondent.
Thereafter, when the AP name of a new AP targeted for management is registered, AP name registration and DVOL allocation are carried out, and when a new DVOL 181 is to be allocated to an existing management-target AP, a DVOL 181 is allocated to this AP. If, for example, the AP administrator 1 wants to know which DVOL 181 is allocated to an AP that he is managing, the AP administrator 1 can use the AP management program 1, and reference the AP management table 133 through the storage management program 131.
An AP 119 records information denoting the execution status of this AP 119 in a log file corresponding to itself (for example, when access to a logical volume began, and when access to a logical volume ended).
When the exact update time recorded in the JVOL 180 should be a time between, for example, after N seconds (N being an integer of 0 or more but less than 60) up to N+1 seconds (when N=59, N+1=60=0) (For example, when a data update occurred after N seconds up to N+1 seconds), the disk control program 191 records the time at N+1 seconds as the update time.
The disk control program 191, either in response to a query from the storage management program 131 or independent of this, actively, either regularly (for example every second) or irregularly, sends to the storage management program 131 difference notification data (for example, the current time, and the VOL identifier of a DVOL 181 for which this time is the update time), which is data showing when an update took place (in other words, when a difference occurred) in which DVOL 181. The storage management program 131 operates the CDP repository 135 each time a prescribed amount of difference notification data (for example, one) is received, and records a difference mark in the location corresponding to the unit of time and VOL identifier specified from the difference notification data. The data update history table 137 is updated regularly in accordance with processing such as this. Using the VOL identifier as a key, it is possible to discern from the data update history table 137 the data update status (that is, at what unit of time a difference occurred) of the DVOL 181 corresponding to this VOL identifier. For example, if explained by referring to
The disk control program 191, in response to a copy indication from a job host 101 or a storage management host 103, can create a copy of a DVOL 181 specified in this copy indication. Since this copy is treated as the base for a restore, hereinafter, this copy will be called the “base VOL”. In a storage system 105, VOL management information, which denotes which VOL is a base VOL, and at what time this base VOL is the base VOL in which DVOL 181, is stored, for example, in the memory 189.
The storage management program 131, for example, receives from the storage administrator 113 a RPO time specification and a VOL identifier of a DVOL 181 targeted for a restore. The storage management program 131, in response to this specification, creates a restore indication comprising the RPO time and VOL identifier, specifies the storage identifier corresponding to this VOL identifier from the DVOL management table 161, and sends this created restore indication to the storage system 105 identified from the specified storage identifier. The restore control program 193 receives this restore indication, and commences restore processing in response to this restore indication. In a restore process, for example, a base VOL, which is the base VOL corresponding to the DVOL (hereinafter called “specified DVOL”) 181 identified from the VOL identifier in the restore indication, and which is the copy of the specified DVOL 181 at a time, which is a previous but recent time from the RPO time (called the “base time” here), is specified. Then, the respective data inside the respective journals of the specified DVOL 181, from after the base time until the RPO time in the specified base VOL, is sequentially written to the base VOL. Thereafter, the VOL identifier of the base VOL and the current VOL identifier of the specified DVOL 181 are swapped (that is, path switching is carried out inside the storage system 105). Consequently, the restore process ends, and the AP 119, which will use the specified DVOL 181, can access the base VOL in which the restored data is stored after the VOL identifier swap. That is, the AP 119 can utilize the restored data.
<<AP Management Support>>
An explanation of support for managing an AP 119 will be given below. A number of cases can be cited for AP 119 management support, but the following three cases will be taken up and explained as typical examples.
(1) First case: A storage failure occurs, and the storage administrator 113 determines the RPO time and restores data irrespective of the AP 119 execution status.
(2) Second case: A storage failure occurs, and the AP administrator 111 determines the RPO time, and makes a restore request to the storage administrator 113.
(3) Third case: Independent of a storage failure occurring, the AP administrator 111 determines the RPO time, and makes a restore request to the storage administrator 113.
The flows of processing for the three cases are covered in
The above-mentioned three cases will be explained hereinbelow by referring to
(First case: A storage failure occurs, and the storage administrator 113 determines the RPO time and restores data irrespective of the AP 119 execution status.)
A storage failure is a malfunction that occurs in a storage system 105. When a storage failure occurs (YES in Step 1001 of
When a storage failure occurs (YES in Step 1021 of
A determination is made as to whether or not the storage administrator 113 will carry out a restore operation (Step 1003 in
The storage administrator 113 sets the RPO time in the storage management program 131. For example, the storage administrator 113 sets the RPO time immediately prior to the failure time (for example, one second prior) to shorten as much as possible the amount of time to restore data. The storage management program 131 creates a restore indication comprising the set RPO time (for example, 19:00:03) and the VOL identifier of the restore-target DVOL 181, and sends this restore indication to the storage system 105 in which the storage failure occurred (Step 1004 of
The storage management program 131 specifies the DVOL (hereinafter, affected DVOL) in which operations are affected by the restore (Step 1005 of
In Step 1006 of
In Step 1007 of
(1) The RPO time specified by the restore indication of Step 1004 (for example, 19:00:03).
(2) The failure time (for example, 19:00:04)
(3) Allocation information, which denotes which DVOL 181 is allocated to which management-target AP 119 of the notification-destination AP management program 117 (for example, the DVOL 181 corresponding to the AP name and VOL identifier specified from the AP management table 133).
(4) The data update status of the respective DVOLs 181 allocated to the respective management-target APs 119 (for example, the DVOLs 181 corresponding to the times and presence or absence of difference marks specified from the data update history table 137).
In Step 1025 of
In Step 1026 of
The AP management support GUI 601A of
The AP management program 1 lines up and displays the data update statuses of volumes 1, 2, 3 and 5 (the differences that occurred at what times to which DVOL), and the execution statuses of APs 1, 2 and 3 (which APs are running at what times). Hereinafter, for the sake of convenience, the area 351 in which the data update statuses are displayed will be called “data update status display area 351”, and the area 353 in which the execution statuses of the APs are displayed will be called “AP execution status display area 353”. The data update status display area 351 and AP execution status display area 353, respectively, can also be GUIs, which are constituted such that the displays are scrollable.
Four time axes (time axes that extend horizontally) corresponding respectively to the four volumes 1, 2, 3 and 5 are prepared in the data update status display area 351, and the four time axes are arranged at intervals in the vertical direction. The respective locations on the thus arranged four time axes are lined up perpendicularly, and signify the same time.
Characters (for example, a VOL identifier and AP name) denoting which DVOL a time axis corresponds to, and which DVOL is allocated to which AP are displayed in the proximity of the respective time axes.
Difference-signifying strip-shaped diagrams (for example, semitransparent diagrams) of a length equivalent to a one second period are stackedly displayed in locations corresponding to one second periods corresponding to difference marks based on the data update statuses of information element (4). If differences occur consecutively for K seconds (K being an integer of no less than 2), the strip-shaped diagrams become continuous, and thus, as a result, the length of the strip-shaped diagrams becomes the length of this K second period.
In the proximity of (for example, immediately below) the respective locations on at least one time axis of the four time axes there is displayed numerals denoting a time equivalent to these locations. Further, of the plurality of times displayed, the time, which is the failure time, is displayed in such a way as to further signify that it is the failure time (for example, the characters “failure” are displayed, or the numerals denoting the time are displayed in bolded form by changing their color or thickness). Further, of the plurality of times displayed, the time, which is the RPO time, is displayed in such a way as to further signify that it is the RPO time (for example, the characters “RPO time” are displayed, or the numerals denoting the time are displayed in bolded form by changing their color or thickness).
Three time axes (time axes that extend horizontally) corresponding respectively to the three APs 1, 2, and 3 are prepared in the AP execution status display area 353, and the three time axes are arranged at intervals in the vertical direction. The vertically aligned respective locations on the thus arranged three time axes and the four time axes arranged in the data update status display area 351 signify the same times.
Characters (for example, an AP name) denoting which AP a time axis corresponds to are displayed in the proximity of the respective time axes.
Operation-signifying strip-shaped diagrams (for example, semitransparent diagrams hereinafter), which signify operation in progress, of a length equivalent to a one second period are stackedly displayed in locations corresponding to one second periods corresponding to an operation in progress based on the execution status information recorded in log files corresponding to the respective APs. If operation is continuous for K seconds (K being an integer of no less than 2), the strip-shaped diagrams also become continuous, and thus, as a result, the length of the strip-shaped diagrams becomes the length of this K second period.
The AP execution status display, for example, can be carried out by the AP execution status display process shown in
The AP administrator 1, who references the above-mentioned AP management support GUI 601A, can specify an AP, which was affected by a data restore resulting from a storage failure. Even if a plurality of APs share one DVOL, it is still possible to specify which AP was used to carry out a data update to a DVOL because the execution statuses of the AP are displayed side-by-side with the data update statuses of the DVOL. From the GUI 601A of
Furthermore, for a volume that is not directly affected by a restore, from a job standpoint, the AP administrator 111 can notify the storage administrator 113 of a restore-target DVOL and a desired RPO time when a restore is deemed necessary.
(Second case: A storage failure occurs, and the AP administrator 111 determines the RPO time, and makes a restore request to the storage administrator 113.)
When a storage failure occurs (YES in Step 1001 of
A determination is made as to whether or not the storage administrator 113 will carry out a restore operation (Step 1003 in
In Step 1012 of
(1) Failure time (for example, 19:00:04),
(2) Allocation information, which denotes which DVOL 181 is allocated to which management-target AP 119 of the notification-destination AP management program 117 (for example, the DVOL 181 corresponding to the AP name and VOL identifier specified from the AP management table 133),
(3) The data update status of the respective DVOLs 181 allocated to the respective management-target APs 119 (for example, the DVOLs 181 corresponding to the times and presence or absence of difference marks specified from the data update history table 137).
In Step 1031 of
In Step 1032 of
In Step 1033 of
The AP management support GUI 601B of
Unlike the GUI 601A of
Looking at this GUI 601B of
Further, since both AP 2 and AP 3 are running when the failure occurs, the AP administrator 1 learns that volumes 1 and 2, which are the affected DVOLs, are being operated by AP 2 and AP 3.
Based on his understanding of these statuses, the AP administrator 1 knows that 19:00:02 is the optimum RPO time. This is because 19:00:02 constitutes the boundary line between the end of operations for AP 1 and AP 2, and the start of operation of AP 3. Furthermore, this status determination can also be carried out by the AP management program 1. That is, based on the data update status from information element (3) and the respective log files of management-targets AP 1, AP 2 and AP 3, the AP management program 1 can carry out the above-described status determination, compute a recommended RPO time, and display the computed RPO time on the above-mentioned GUI 601B. This can also be applied in the same way to the third case, which will be explained hereinbelow.
The preceding is an explanation of the second case. Furthermore, details of the processing for displaying the GUI can be made the same as the processing of the first case. This also holds true for the following third case.
(Third case: Independent of a storage failure occurring, the AP administrator 111 determines the RPO time, and makes a restore request to the storage administrator 113.)
When Step 1021 of
When Step 1001 of
The AP management program 117 transitions to the waiting-to-receive mode subsequent to Step 1027 of
The AP management support GUI 601C of
Accordingly, the AP administrator 2 can select 19:00:02 as the RPO time, and can input to the AP management program 2 a request to restore data in volume 3 and volume 4 at this RPO time (19:00:02) with the understanding that it will not impede the job and will not adversely affect another job host if only the AP 4 job, which he is managing himself, is restarted. In response to this input, the AP management program 2 sends to the storage management program 131 a restore request comprising volumes 3 and 4, and an RPO time of 19:00:02. The storage management program 131 can respond to this restore request by sending to the storage system 105 a restore indication having an RPO time (19:00:02) and volumes 3 and 4 as restore targets.
Furthermore, in the example of
Further, the AP management program 2 can refuse to accept an RPO time for a time at which a difference has occurred due to the operation of an AP of another job host. For example, the AP management program 2 can display the message “Cannot be selected” when 19:00:01 is selected as the RPO time. Furthermore, in this case, the fact that a difference occurred due to an operation in the other job host 1 can be displayed as shown in
<Second Embodiment>
A second embodiment of the present invention will be explained hereinbelow. Hereinafter, the points of difference with the first embodiment will mainly be explained, and explanations of those points that the second embodiment has in common with the first embodiment will be simplified or omitted.
In the second embodiment, no less than two DVOLs of a plurality of DVOLs are managed as a single volume group.
Which DVOLs belong to which volume group is also registered in the DVOL management table 161′. For example, volume group names corresponding to the VOL identifiers of the respective DVOLs are registered. A volume group can be constituted from no less than two DVOL 181 in one storage system, or a volume group can be constituted from no less than two DVOLs 181 residing in a plurality of storage systems (That is, a volume group can span a plurality of storage systems.). For example, volume group 3 spans storage systems 1 and 2. In this case, for example, when the storage management program 131 indicates a restore due to a storage failure in storage system 1, since volume 1 inside storage system 1 makes up volume group 3 together with volume 6 of storage system 2, the storage management program 131 can send, also for volume 6, a restore indication specifying the same RPO time to storage system 2.
The AP management support GUI 601 displays which DVOLs belong to which volume groups. For example, the group names of volume groups of DVOLs corresponding to VOL identifiers are displayed near these VOL identifiers, which are written in the proximity of the respective time axes in the data update status display area 351. The AP administrator 111 can also take into account which DVOLs belong to which volume groups when sending a restore request to the AP management program 117. More specifically, for example, when the AP administrator 111 wants to make volume 4 a restore target, because volume 3 belongs to the same volume group as this volume 4, the AP administrator 111 can specify to the AP management program 117 to carry out a restore that has as the restore target volume 3 as well as volume 4.
A number of embodiments of the present invention have been explained hereinabove, but these are examples for explaining the present invention, and do not purport to limit the scope of the present invention solely to these embodiments. The present invention can be put into practice in a variety of other modes. For example, the host computer can also send a restore indication to a storage system. Further, for example, the present invention is not limited to graph format displays such as those shown in
Number | Date | Country | Kind |
---|---|---|---|
2007-098052 | Apr 2007 | JP | national |
Number | Name | Date | Kind |
---|---|---|---|
7111136 | Yamagami | Sep 2006 | B2 |
7143413 | McCarthy et al. | Nov 2006 | B2 |
7162601 | Yamagami | Jan 2007 | B2 |
7263590 | Todd et al. | Aug 2007 | B1 |
7383465 | van Rietschote et al. | Jun 2008 | B1 |
7398422 | Amano et al. | Jul 2008 | B2 |
7401248 | Nakahara et al. | Jul 2008 | B2 |
7469358 | Kaiya et al. | Dec 2008 | B2 |
7475206 | Murotani et al. | Jan 2009 | B2 |
7660958 | Shackelford et al. | Feb 2010 | B2 |
7681001 | Maruyama et al. | Mar 2010 | B2 |
7725445 | Hirakawa et al. | May 2010 | B2 |
8086808 | Ichikawa | Dec 2011 | B2 |
20010054133 | Murotani et al. | Dec 2001 | A1 |
20060195493 | Chang et al. | Aug 2006 | A1 |
20070050545 | Tsuboki et al. | Mar 2007 | A1 |
20070271475 | Hatasaki et al. | Nov 2007 | A1 |
20070277011 | Tanaka et al. | Nov 2007 | A1 |
20080082777 | Sakaguchi et al. | Apr 2008 | A1 |
20080120352 | Nishi et al. | May 2008 | A1 |
Number | Date | Country |
---|---|---|
2005-018738 | Jan 2005 | JP |
Number | Date | Country | |
---|---|---|---|
20080250417 A1 | Oct 2008 | US |