JP2016224864A - Storage system migration method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To ensure the safe migration of data in a storage system.SOLUTION: A method of migrating a storage system having an administration server and storing original data and duplicated data of the original data in different storage devices, respectively, comprises: calculating, by the administration server, an idle capacity of each storage device of the storage system; identifying, by the administration server, the storage device the idle capacity of which is equal to or larger than a capacity of the duplicated data as a candidate; identifying, by the administration server, a combination of the storage device identified as the candidate and a storage device storing the original data; calculating, by the administration server, a sum of values of data protection weights of the identified combination; and registering, by the administration server, the identified combination as a relay duplication destination when the calculated sum exceeds a predetermined value.SELECTED DRAWING: Figure 6

Description

  The present invention relates to a storage system migration method and program.

  Regarding data protection of storage systems, Patent Document 1 discloses a quantitative assessment of data loss risk against disasters such as tsunamis for storage devices installed at geographically dispersed sites, and finds low-risk sites. A technique for selectively replicating data to a disaster-safe site is disclosed.

US Published Patent 2013/0290623

  If the technique disclosed in Patent Document 1 is used, a safe site can be specified as a data replication destination. On the other hand, the storage system can find a safer site after the operation is started, for example, when a safe site is additionally installed, or when the storage capacity of the safe site is expanded, It is desirable to migrate replicated data to that site.

  On the other hand, there is a possibility that a disaster may occur in the middle of the migration, and Patent Document 1 does not disclose a technology relating to the secure migration of replicated data.

  Therefore, an object of the present invention is to safely migrate storage system data.

  In order to solve the above-described problem, a storage system migration method according to the present invention includes a management server, and stores the original data and the duplicate data of the original data in different storage apparatuses. The management server calculates a free capacity of a storage device of the storage system, specifies a storage device having the calculated free capacity equal to or larger than the capacity of the replicated data as a candidate, and the storage device specified as the candidate; A combination with the storage device in which the original data is stored is identified, a sum of the data protection weight values of the identified combination is calculated, and if the calculated sum exceeds a predetermined value, the identified combination is It is registered as a relay duplication destination.

  According to the present invention, data in a storage system can be safely migrated.

It is a figure which shows the example of a structure of a storage system. It is a figure which shows the example of a storage structure management table. It is a figure which shows the example of a data protection weight management table. It is a figure which shows the example of a replication structure management table. It is a flowchart which shows the example of a replication destination determination program. It is a flowchart which shows the example of a duplication plan program. 10 is a diagram illustrating an example of a replication configuration management table in Embodiment 2. FIG. 12 is a flowchart illustrating an example of a capacity adjustment / duplication plan program according to the second embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following description, various types of information may be described using an expression such as “management table”, but the various types of information may be expressed using a data structure other than a table. Further, the “management table” can be referred to as “management information” to indicate that it does not depend on the data structure.

  Further, the process may be described with “program” as the subject. The program is executed by a processor, for example, an MP (Micro Processor) or a CPU (Central Processing Unit), and performs a predetermined process. Since processing is performed using storage resources (for example, a memory) and a communication interface device (for example, a communication port) as appropriate, the subject of the processing may be a processor.

  The processor may have dedicated hardware in addition to the MP and CPU. The computer program may be installed on each computer from a program source. The program source may be provided by, for example, a program distribution server or a storage medium.

  Each element such as a table can be identified by a number or the like, but other types of identification information such as a name may be used in addition to the number as long as the information can be identified. In the following description and drawings, the same reference numerals are given to the same parts, but the present invention is not limited to the present embodiment, and any application examples that meet the idea of the present invention are within the technical scope of the present invention. include. Further, unless specifically limited, each component may be plural or singular.

<Generate relay duplication destination>
A first embodiment for carrying out the present invention will be described below with reference to FIGS. In the first embodiment, one storage is newly added to a storage system composed of three storages that have already stored replica (replica) data, and the duplication destinations of the four storages are reviewed. To be migrated.

  FIG. 1 is a block diagram illustrating an example of the configuration of the storage system 100. The storage system 100 includes three or more storages 101a to 101n (hereinafter, an arbitrary one is represented as the storage 101) and a management server 102. The management server 102 includes a CPU 106, a memory 107, an interface 108 (hereinafter referred to as an operation I / F 108) for connecting to the management network 105, and an interface 109 (hereinafter referred to as console I / F) for connecting to the management console 103. F109).

  The memory 107 includes a storage configuration management table 110, a data protection weight management table 111, a replication configuration management table 112, a replication destination determination program 113, a replication plan program 114, a replication control program 115, a capacity adjustment / duplication plan. A program 116 is stored, and the CPU 106 executes these programs. The capacity adjustment / duplication plan program 116 will be described in the second embodiment.

  The management network 105 is a network for the management server 102 to manage the storage 101, and Ethernet (registered trademark) or the like is preferable. The data network 104 is a network for transferring data between the plurality of storages 101, and Ethernet, fiber channel, the Internet, etc. are preferable. The data network 104 may be the same network as the management network 105.

  In order to specify each of the storages 101a to 101n, an ID is assigned to each storage 101. In the example of FIG. 1, an ID “storage A” is assigned to the storage 101a. In the following, the storage 101 a assigned with the ID “storage A” is represented, and therefore it may be referred to as “storage A” for convenience.

  The storage 101 includes a CPU (not shown), a memory, a volume for storing data, an interface (operation I / F) for connecting to the management network 105, and an interface (data I / F) for connecting to the data network 104. / F).

  A replication program is arranged in the memory included in the storage 101, and the CPU included in the storage 101 executes the replication program. The replication program communicates with the replication program of the other storage 101 via the data network 104 and replicates the data stored in the volume to the other storage 101. Data replication units are volume units, block units, file units, object units, and the like.

  The table and each program stored in the memory 107 may be stored in the memory included in the storage 101, and the CPU included in the storage 101 may execute the stored program.

  Hereinafter, when the storage system 100 is configured with three storages 101 and the data redundancy of each storage 101 is 2 (that is, when one copy of data stored in each storage 101 has been created one by one) The operation of migrating data to a new copy destination obtained by adding one new storage 101 will be described.

  FIG. 2 is a diagram illustrating an example of the storage configuration management table 110. The storage configuration management table 110 stores at least an ID 200 for specifying the storage 101, a data amount 201 stored in each storage 101, a free capacity 202 for storing the replicated data in each storage 101, and each storage 101. It has data redundancy 203 and an IP address 204 necessary for accessing each storage 101.

  In the example of FIG. 2, “storage A”, “storage B”, “storage C”, and “storage D” are used as the ID 200 information, but information of other expressions may be used. Further, the ID 200 may be information specified not in units of the storage 101 but in units of volumes.

  Furthermore, although the IP address 204 shows the IP address of the operation I / F that each storage 101 has as an example, depending on the configuration of the network, for example, it may be the data I / F that each storage 101 has, Both may be included. Further, when the network is configured with a fiber channel, a WWN (World Wide Name) may be included. The storage management table 110 is created by the administrator using the management console 103 prior to duplication described below.

  FIG. 3 is a diagram illustrating an example of the data protection weight management table 111. The data protection weight management table 111 includes at least a replication source 300, a replication destination 301, and a data protection weight 302 that is a value for each of these combinations.

  The data protection weight 302 is a numerical value indicating the strength of data protection against a disaster. The data protection weight 302 is calculated from the disaster risk of each storage 101 and the data amount of each storage 101. The disaster risk is a probability that the storage 101 of the replication source 300 and the replication destination 301 is damaged at the same time due to a disaster such as an earthquake or a tsunami, and the stored data is lost.

  At this time, the data protection weight 302 of the replication performed by the storages A to D of the replication source 300 to the storages A to D of the replication destination 301 indicates the probability of losing data between the storages A to D from the value “1”. It is obtained by multiplying the subtracted value by the amount of data stored in the storages A to D of the copy source 300. Here, instead of the amount of data stored in the storages A to D of the replication source 300, the size of the volume in which the data of the storages A to D of the replication source 300 may be stored may be the volume size. It is good also as capacity independent of.

  In the example of FIG. 3, when data is copied from the storage A to the storage B, the probability of losing the data is “0.1”, and the data amount of the storage A is “1”. Is “0.9”. Further, when data is copied from the storage B to the storage C, the probability of losing the data is “0.2”, and the data amount of the storage B is “1”, so the data protection weight is “0. 8 ". Since the replication from the storage A to the storage A is meaningless from the viewpoint of data protection, the value of the data protection weight is “−” without definition.

  It is assumed that the data protection weight management table 111 is created and set in advance before executing replication. The disaster risk necessary for the creation of the data protection weight management table 111 may be calculated, for example, from the statistical information on actual earthquake or tsunami damage, or using existing earthquake models or tsunami models. It may be calculated or calculated using a numerical simulator. The data protection weight management table 111 may be set from the management console 103 based on the calculated disaster risk.

  The items of the replication source 300 and the items of the replication source 301 may include all of the storage 101 or only a part thereof.

  FIG. 4 is a diagram illustrating an example of the replication configuration management table 112. The replication configuration management table 112 includes at least a replication source 400 that is the ID of the replication source storage 101, a current replication destination 401 that indicates the ID of the storage 101 that is the current replication destination of the stored data, and the next of the storage data. A relay duplication destination 402 indicating the ID of the copy destination storage 101 and a migration duplication destination 404 indicating the ID of the storage 101 where the copy of the stored data is to be finally stored. As shown in FIG. 4, the relay replication destination 402 may constitute a plurality of fields such as the relay replication destination 402a and the relay replication destination 402b.

  After the storage system 100 determines the replication destination, the management console 103 can display the replication configuration management table 113 and accept corrections. For example, the administrator can operate the management console 103 to check whether there is a problem in the duplication destination or to adjust according to special requirements.

  In the example of FIG. 4, as the replication source 400, the current replication destination 401, the relay replication destination 402, and the migration replication destination 403, the same information as the ID 200 of the storage 101 is included, but the IP address of the storage 101 204 may be sufficient. In the example of FIG. 4, only two sets (two columns) of relay duplication destinations 402 are included, but a plurality of sets of three sets (three columns) or more may be included.

  Here, the present is the time when a predetermined redundancy is ensured by copying from the migration source 400 to the first replication destination, and the current replication destination 401 determined by the replication destination determination program 113 described below. It is the time when it was copied to. The replication destination determination program 113 performs a replication planning process for data migration and a replication destination determination process for determining a migration replication destination before executing data replication using the replication planning process.

  FIG. 5 is a flowchart illustrating an example of a procedure for determining the current replication destination 401 and the migration replication destination 403 in the replication configuration management table 112. Since the procedure for determining the current replication destination 401 and the procedure for determining the migration replication destination 403 are the same, the replication destination described with reference to FIG. 5 is either the current replication destination 401 or the migration replication destination 403. is there.

  When the replication destination determination process is started, the replication destination determination program 113 first reads the data protection weight management table 111 (S501). Next, the sum of the data protection weights is calculated for all possible combinations. (S502). Here, “executable” means that the data is copied so that all the storages 101 have the same number of data as the redundancy 203 in the system without running out of the free capacity 202 of the storage configuration management table 110. is there.

  Finally, the storage combination ID of the best combination is written in the replication destination of the replication configuration management table 112 (S503), and the replication destination determination process for determining the replication destination ends. Here, the best combination indicates a combination that maximizes the sum of data protection weights. In order to determine the current replication destination 401 and the migration replication destination 403, the above procedure is executed once each.

  In this example, when the current replication destination 401 is determined, only the storage 101 whose IDs are storage A, storage B, and storage C among the storage 101 is registered in the system. Therefore, the storage D does not exist in the data protection weight management table 111 read in S501, and neither the copy source nor the copy destination.

  In S502, the copy destination determination program 113 obtains two executable combinations. The first combination is a combination in which data is copied from storage A to storage B, storage B to storage C, and storage C to storage A. The sum of the data protection weights at this time is 0.9 + 0.8 + 0.8 = 2.5. The second combination is a combination in which data is copied from storage A to storage C, storage B to storage A, and storage C to storage B. The sum of the data protection weights at this time is 0.5 + 0.9 + 0.5 = 1.9.

  In S <b> 503, the replication destination determination program 113 determines that the first combination that is a combination with a large sum of data protection weights (that is, a safe) is the best, and writes the first combination in the current replication destination 401 of the replication configuration management table 112.

  Further, in this example, when the migration replication destination 403 is determined, the storage 101 whose IDs are storage A, storage B, storage C, and storage D is registered in the system. Therefore, in S502, the replication destination determination program 113 obtains nine executable combinations and calculates the sum of data protection weights for each combination. In S503, the copy destination determination program 113 determines that the combination of copying data (storage data sum of weights) is that storage A is storage B, storage B is storage A, storage C is storage D, and storage D is storage C. Is the best), and is written in the migration replication destination 403 of the replication configuration management table 112.

  FIG. 6 is a flowchart illustrating an example of a procedure for determining the relay replication destination 402 of the replication configuration management table 112. When the replication planning process is started, the replication planning program 114 sets the current replication destination 401 as a processing target (S601). In step S <b> 601, the replication plan program 114 reads the ID of the current replication destination 401 in the replication configuration management table 112. Here, the processing target is the current replication destination 401. However, since the processing target is sequentially changed in the processing described below, it is simply described as the processing target replication destination.

  Next, the replication planning program 114 creates a replication destination combination A that approaches the migration replication destination 403 (S602). In S602, the replication planning program 114 first calculates the free capacity of each storage 101 when replication data is stored in the replication destination to be processed. Next, the replication planning program 114 uses the free capacity to create a combination candidate that reduces the difference between the processing target replication destination and the migration replication destination 403.

  Note that the replication destination approaching the migration replication destination 403, or the difference between the target replication destination and the migration replication destination 403 is small, the processing target replication destination storage 101 and the storage replication destination 403 storage 101 are the same. The number of storages 101 of the migration source 400 is large.

  For example, when the processing destination replication destination is the current replication destination 401, as shown in FIG. 4, the storage B of the current replication destination 401 and the storage B of the migration replication destination 403 are compared to the storage A of the migration source 400. The storage C of the current copy destination 401 and the storage A of the transfer destination copy destination 403 are different from the storage B of the transfer source 400. For this reason, when only the two storage units A and B of the migration source 400 are viewed, the number of storages 101 of the migration source 400 in which the storage destination copy destination storage 101 and the storage copy destination 403 storage 101 are the same. Is one. A state where the number of storages 101 such as the storage A is large is a state where the difference is small.

  Then, the replication planning program 114 reads the storage configuration management table 110 and detects that the storage D has a free capacity of the value “1” in a state where the replication data is stored in the current replication destination 401. Next, the replication planning program 114 refers to the migration replication destination 403 and adds the storage D as the replication destination of the storage C, so that the difference between the current replication destination 401 and the migration replication destination 403 can be reduced. Get A.

  Here, the obtained combination A is a combination in which data is copied to the storage A, the storage B, the storage B to the storage C, and the storage C to the previously copied storage A and the newly copied storage D, respectively. It is.

  Next, the replication planning program 114 creates a replication destination combination B from which redundant data exceeding the redundancy 203 value is deleted from the combination A (S603). In S603, the replication planning program 114 sets a combination obtained by deleting a previously determined replication destination of the replication source to which a replication destination is newly added in S602 as a combination B.

  For example, the replication destination to be processed is a combination in which storage A is storage B, storage B is storage C, storage C is storage A, and the combination A is storage A is storage B. When the storage B is a combination that replicates data to the storage C and the storage C is the storage A and the storage D, respectively, the fact that the storage C of the combination A is replicated to the storage A is included in the processing target replication destination Therefore, the combination B is a combination in which the storage A is copied to the storage B, the storage B is copied to the storage C, and the storage C is copied to the storage D, respectively.

  If there is no redundant data exceeding the value of the redundancy 203 in the combination A, the replication planning program 114 does not delete the redundant data.

  The replication plan program 114 calculates the sum of the data protection weights of the combination B (S604). In S604, the replication planning program 114 reads the data protection weight management table 111 and calculates the sum of the data protection weights of the combination B. For example, in the data protection weight management table 111 shown in FIG. 3, the value of each data protection weight is 0.9 when the storage A replicates to the storage B, and 0.8 when the storage B replicates to the storage C. Since the case where the storage C replicates to the storage D is 0.9, the sum is 0.9 + 0.8 + 0.9 = 2.6.

  The replication planning program 114 determines whether or not the sum of the data protection weights of the combination B is greater than the sum of the data protection weights of the current replication destination 401 (S605). In S605, the replication plan program 114 compares the sum of the data protection weights of the combination B obtained in S604 with the sum of the data protection weights of the current replication destination 401, and the combination B has increased. Proceeds to S606, otherwise proceeds to S602.

  For example, if the current replication destination 401 is a combination that replicates data to storage B, storage B to storage C, and storage C to storage A as shown in FIG. From the value of the data protection weight management table shown, the sum of the data protection weights is 0.9 + 0.8 + 0.8 = 2.5. For this reason, when the sum of the data protection weights of the combination B is 2.6 as described above, the replication planning program 114 proceeds to S606.

  If the sum of the data protection weights of the combination B is smaller than 2.5, the replication planning program 114 proceeds to S602 and creates another combination A.

  In step S606, the replication plan program 114 determines whether the combination B matches the combination of the storage 101 of the migration replication destination 403. When the combination B matches the combination of the storage 101 of the migration replication destination 403, the replication plan process is terminated. Otherwise, the replication plan program 114 proceeds to S607.

  In step S <b> 607, the replication plan program 114 writes the combination B in the relay replication destination 402 of the replication configuration management table 112. In step S607, the replication plan program 114 writes the combination B to the relay replication destination 402a if the processing target is the current replication destination 401, and writes the combination B to the relay replication destination 402b if the processing target is the relay replication destination 402a. . Here, a relay replication destination after the relay replication destination 402b may exist.

  Next, the replication planning program 114 sets a new relay replication destination as a processing target (S608). In S608, the replication planning program 114 sets the relay replication destination written in the replication configuration management table 112 in S607 as the next processing target, and proceeds to S602.

  As is clear from the above description, the replication planning program 114 repeats the processing from S602 to S606 until a relay replication destination 402 that matches the migration replication destination 403 is obtained. Thereafter, the replication control program 115 refers to the replication configuration management table 113 and instructs the replication program of each storage 101 to replicate. At this time, the relay duplication destinations 402 are instructed one by one in order, and the duplication program of each storage 101 gives the next instruction after completing the duplication of data.

  For example, first, when the replication control program 115 instructs replication to each storage 101 written in the relay replication destination 402 a, the replication program of each storage 101 instructed will be stored in each storage 101 written in the current replication destination 401. The data is replicated so that the storages 101 written to the relay replication destination 402a are stored.

  Next, when the replication control program 115 instructs replication to each storage 101 written to the relay replication destination 402b, the replication program of each storage 101 replicates data to each storage 101 written to the relay replication destination 402b. Do. Finally, when the replication control program 115 instructs replication to each storage 101 written to the migration replication destination 403, the replication program of each storage 101 performs replication to each storage 101 written to the relay replication destination 402b, Complete replication.

  In the first embodiment, the case where the data amount is 1, the free capacity is 1, and the redundancy is 2 has been described as an example. However, when the number of copies is the data amount and the free capacity is 2 or the redundancy is 3 or more. A similar process can be used. In the first embodiment, the data protection weight is calculated from the disaster risk and the data amount of each storage 101. However, the data protection weight may be obtained by simply subtracting the disaster risk from the value “1”.

  Further, in the first embodiment, the replication control program 115 creates one combination A, creates a corresponding combination B in order to create one relay duplication destination 402, and calculates the sum of the data protection weights. Then, one of the relay duplication destinations 402 is selected by determining whether the sum is larger than the current duplication destination 401. However, instead of such processing, a plurality of combinations A are created at one time. Also good.

  That is, the replication control program 115 creates a plurality of combinations A, creates a combination B corresponding to each combination, calculates the sum of the data protection weights, selects the combination B having the largest sum, It may be determined whether the data protection weight sum of the selected combination B is larger than the current copy destination.

  In the first embodiment, the replication control program 115 instructs the replication to each storage 101 written to the relay replication destination 402 one by one, and waits for the completion of the replication program each time. The program 115 instructs the duplication to the two or more storages 101 written in the relay duplication destination 402, instructs the duplication to the two or more storages 101 written in the migration duplication destination 403, and the duplication program May be executed one by one, and during that time, the replication control program 115 does not have to wait for completion.

  As described above, the replicated data can be safely migrated from the current replication destination 401 to the migration replication destination 403 without lowering the sum of data protection weight values, that is, via the relay replication destination 402 safely.

<Generate relay duplication destination when there is no free space>
Hereinafter, Example 2 which is another form for implementing this invention is demonstrated with reference to FIG. 7 and FIG. In the second embodiment, description of portions that are not different from the processing in the first embodiment is omitted, and only the difference is described.

  In the second embodiment, even when there is no free space in each storage 101 due to the duplicated data in the first embodiment, a relay duplication destination is generated and can be transferred to the target duplication destination. In the second embodiment, the storage system 100 secures a free capacity by eliminating the replication destination according to the data protection weight before the stage of generating the relay replication destination described in the first embodiment.

  In the second embodiment, the capacity adjustment / replication planning program 116 is used in addition to the replication planning program 114 in the first embodiment, and further, tolerances not shown are stored in the memory 107 of the management server 102. The tolerance is a value corresponding to the sum of the data protection weights calculated for each combination of the storages 101, and the management console 103 is operated by the administrator prior to the execution of the capacity adjustment / replication planning program 116. Is set. The following description is an example in which “2.5” is set as the tolerance.

  FIG. 7 is a diagram illustrating an example of the replication configuration management table 112 according to the second embodiment. The replication configuration management table 112 includes at least a replication source 700 that indicates the ID of the storage 101 that is the replication source, a current replication destination 701 that indicates the ID of the storage 101 that is the current replication destination of the stored data, and the next of the storage data. A relay duplication destination 702 indicating the ID of the copy destination storage 101 and a migration duplication destination 704 indicating the ID of the storage 101 where the copy of the stored data is to be finally stored.

  As shown in FIG. 7, the relay duplication destination 702 may constitute a plurality of fields such as the relay duplication destination 702a and the relay duplication destination 702b. The difference between the second embodiment and the first embodiment in the replication configuration management table 112 is the ID value of the storage destination storage 101. The ID of the storage 101 of the current copy destination 701 is input by the operation of the administrator, and the ID of the storage copy destination 703 storage 101 is set by the copy destination determination program 113. The relay replication destination 702 is set during execution of a capacity adjustment / replication destination replication planning program 116 described later. The items in the replication configuration management table 112 are the same as those in FIG. 4, but since the ID value of the storage 101 is different, a different code is attached to indicate the ID of the storage 101.

  FIG. 8 is a flowchart illustrating an example of a procedure for determining the relay replication destination 702 of the replication configuration management table 112 according to the second embodiment. When the capacity adjustment / replication planning process is started, the capacity adjustment / replication planning program 116 targets the current replication destination 701 (S801). In S801, the capacity adjustment / replication plan program 116 sets the copy destination written in the current copy destination 701 of the copy configuration management table 112 as a processing target.

  Next, the capacity adjustment / duplication plan program 116 obtains a combination C in which some duplication destinations have been eliminated (S802). In S802, the capacity adjustment / duplication plan program 116 reads the data protection weight management table 111, and the replication source 700 is selected from the replication relationships that are different from the migration replication destination 703 in the current replication destination 701. The combination of the replication source 700 and the replication destination having the lowest data protection weight in the relationship of the current replication destination 701 is searched, and the replication destination is canceled from the replication relationship.

  For example, as shown in FIG. 7, the current replication destination 701 from the replication source 700 is a combination of storage A to storage B, storage B to storage C, storage C to storage D, and storage D to storage A. Yes, if the replication source 703 from the replication source 700 is a combination of storage A to storage B, storage B to storage A, storage C to storage D, and storage D to storage C, capacity adjustment / duplication plan The program 116 compares the data protection weight (0.8) of the replication relationship from the storage B to the storage C with the data protection weight (0.5) of the replication relationship from the storage D to the storage A, and performs data protection. The replication destination of the replication relationship from the storage D with the low weight to the storage A is eliminated. Therefore, the combination C is a combination for copying data from the storage A to the storage B, from the storage B to the storage C, and from the storage C to the storage D.

  Next, the capacity adjustment / duplication plan program 116 determines whether the combination C is allowed (S803). If it is determined that it is allowed, the capacity adjustment / replication planning program 116 proceeds to S804, and if it is determined that it is not permitted, the process proceeds to S802 to obtain a combination C in which another replication destination is eliminated. Here, the combination C being allowed means that the sum of the data protection weights of the combination C is larger than the tolerance value. For example, since the data protection weight of the combination C described with reference to FIG. 7 is 0.9 + 0.8 + 0.9 = 2.6, the data protection weight is larger than the tolerance value 2.5, which is permitted in this example.

  In step S804, the capacity adjustment / replication plan program 116 executes a replication plan process. The replication plan process is a process performed by the replication plan program 114 described with reference to FIG. When the replication plan process of S804 is completed, the capacity adjustment / replication plan process is ended.

  As described above, the relay replication destination 702 can be generated even in the storage system 100 in which the free space has been lost due to the storage of the replication data. As a result, it is possible to perform data migration without expanding the capacity of the storage 101 while ensuring data safety.

  In the second embodiment, the tolerance is a value corresponding to the sum of the data protection weights calculated for each combination of the storages 101. However, the tolerance is 10% of the sum of the data protection weights of the current copy destination 701. It is good also as a ratio. In this case, prior to the execution of the capacity adjustment / replication plan program 116, the management server 102 calculates the data from the total value of the data protection weights obtained from the current replication destination 701 and a set ratio such as 10%. By calculating a value corresponding to the sum of the protection weights, the capacity adjustment / replication planning program 116 can be executed in the same manner as described above.

  Further, in the second embodiment, the procedure for determining whether or not the combination C is permitted every time one combination C is obtained. However, after all the combinations C included in the relay duplication destination 702 are obtained, all combinations C are permitted. It may be a procedure for determining whether the degree is exceeded.

  In addition to the first and second embodiments described above, together with the data protection weight as a criterion for determining the relay duplication destinations 402 and 702, migration data of the entire storage system 100 expected when migrating to the relay duplication destinations 402 and 702 An upper limit of the amount may be set in advance, and each storage 101 of the relay duplication destinations 402 and 702 may be selected from a combination that satisfies the upper limit.

  This makes it possible to estimate the time required to move to one relay duplication destination 402, 702. Therefore, when transferring a large amount of data that needs to be transferred over multiple days, the amount of data transferred per day It is possible to select the storage 101 of the appropriate relay replication destinations 402 and 702, and it is possible to easily create a data migration plan. Note that the upper limit of the migration data amount of each storage 101 may be set instead of the upper limit of the migration data amount of the entire storage system 100.

  In addition, data such as volumes and files of one storage 101 may be divided and stored in the plurality of storages 101 at the relay duplication destinations 402 and 702, respectively. For this purpose, the management server 102 may be configured to manage a plurality of replication destinations for one storage 101 and further record data stored in each replication destination. Here, the file names may be recorded one by one, or a plurality of files and folders may be recorded together by a folder name or the like.

  Further, instead of determining the relay replication destinations 402 and 702 one by one until the management server 102 matches the migration replication destinations 403 and 703, the management server 102 determines the relay replication destination 402 that matches the migration replication destinations 403 and 703. , 702 columns may be calculated as candidates, and a combination column that provides the highest average data protection weight of the entire columns of the relay duplication destinations 402 and 702 may be obtained. As a result, it is possible to obtain a sequence of relay duplication destinations 402 and 702 such that each of the plurality of relay duplication destinations 402 and 702 is secure on average.

100 Storage System 101 Storage 102 Management Server 103 Management Console 110 Storage Configuration Management Table 111 Data Protection Weight Management Table 112 Replication Configuration Management Table 113 Replication Destination Determination Program 114 Replication Planning Program 115 Replication Control Program

Claims (15)

A storage system migration method that includes a management server and stores original data and duplicate data of the original data in different storage devices,
The management server is
Calculate the free capacity of the storage device of the storage system,
Identify the storage device whose calculated free space is greater than or equal to the capacity of the replicated data as a candidate,
A combination of the storage device identified as the candidate and the storage device in which the original data is stored;
Calculating the sum of the data protection weight values of the identified combinations;
When the calculated total exceeds a predetermined value, the specified combination is registered as a relay duplication destination. The combination of the storage device identified as the candidate and the storage device in which the original data is stored is characterized by identifying the combination in which the duplicate data is deleted from the storage device that has already stored the duplicate data. The storage system migration method according to claim 1. The management server is
The combination of the storage device storing the original data and the storage device storing the replicated data before the addition of a new storage device, where the sum of data protection weight values is the maximum value Is registered as a copy destination of, and the maximum value is set as the predetermined value,
A combination of the storage device storing the original data and the storage device storing the replicated data after the addition of the new storage device, wherein the sum of the data protection weight values is the maximum value 3. The storage system migration method according to claim 2, wherein the migration system is registered as a migration replication destination. The combination of the storage device specified as the candidate and the storage device in which the original data is stored is specified by specifying the migration copy destination storage device different from the current copy destination as the candidate. The storage system migration method according to claim 3, wherein the storage system migration method is included in an apparatus. The combination of the storage device specified as the candidate and the storage device in which the original data is stored is specified by specifying the migration replication destination storage device different from the registered relay replication destination as the candidate. The storage system migration method according to claim 4, wherein the storage system migration method is included in a storage device. 6. The storage system migration method according to claim 5, wherein when the registered relay migration destination storage device and the migration replication destination storage device are the same, the replication planning process is terminated. The identification information of the storage device in which the original data is stored, the current replication destination storage device, the relay replication destination storage device, and the migration replication destination storage device is used as a result of the replication planning process. The storage system migration method according to claim 6, wherein: The management server is
Instructing the storage apparatus having the identification information resulting from the replication planning process,
The storage system migration method according to claim 7, wherein the storage apparatus instructed to replicate executes replication. 9. The storage system according to claim 8, wherein the value of the data protection weight is a value that is included in the storage system and is calculated based on a probability that a failure occurs simultaneously in a plurality of different storage apparatuses. Migration method. The management server
The tolerance is preset,
Calculate the sum of the values of the data protection weights for the combination of the storage device in which the original data is stored and the storage device with the smallest value of the data protection weights that eliminate the replication relationship of the replicated data, The storage system migration method according to claim 9, wherein when the size is large, the replication relationship is canceled to generate a free capacity of the storage device. A computer program for controlling a storage system for storing original data and duplicated data of the original data in different storage devices,
The computer is
Calculate the free capacity of the storage device of the storage system,
Identify the storage device whose calculated free space is greater than or equal to the capacity of the replicated data as a candidate,
A combination of the storage device identified as the candidate and the storage device in which the original data is stored;
Calculating the sum of the data protection weight values of the identified combinations;
When the calculated sum exceeds a predetermined value, the specified combination is registered as a relay duplication destination. The combination of the storage device identified as the candidate and the storage device in which the original data is stored is characterized by identifying the combination in which the duplicate data is deleted from the storage device that has already stored the duplicate data. The program according to claim 11. The computer
The combination of the storage device storing the original data and the storage device storing the replicated data before the addition of a new storage device, where the sum of data protection weight values is the maximum value Is registered as a copy destination of, and the maximum value is set as the predetermined value,
A combination of the storage device storing the original data and the storage device storing the replicated data after the addition of the new storage device, wherein the sum of the data protection weight values is the maximum value The program according to claim 12, wherein the program is registered as a migration copy destination. The combination of the storage device specified as the candidate and the storage device in which the original data is stored is specified by specifying the migration copy destination storage device different from the current copy destination as the candidate. The program according to claim 13, wherein the program is included in an apparatus. The combination of the storage device specified as the candidate and the storage device in which the original data is stored is specified by specifying the migration replication destination storage device different from the registered relay replication destination as the candidate. The program according to claim 14, wherein the program is included in a storage device.

Images