JP2008123529A - Storage system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To transfer or duplicate data between a plurality of data storage systems without exerting an influence on higher-order computers of the storage systems and to prevent communication between the storage system and the computer from being affected. <P>SOLUTION: In a first storage system, information about updating of data stored in the first storage system is stored as a journal. Specifically, the journal comprises the copy of data used for the updating and update information such as a write instruction at updating time. A second storage system acquires the journal via a communication line between the first storage system and the second storage system. The second storage system holds the copy of data held in the first storage system and uses the journal to update data corresponding to data of the first storage system in the order of data update in the first storage system. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a storage system, and more particularly to data replication between a plurality of storage systems.

  In recent years, technology to replicate data between storage systems is important so that the data processing system can provide services even if a failure occurs in the first storage system in order to always provide continuous service to customers. It has become. As a technique for copying information stored in the first storage system to the second and third storage systems, there is a technique disclosed in the following patent document.

  In US Pat. No. 5,170,480, a first computer connected to a first storage system transfers data stored in the first storage system via a communication link between the first computer and the second computer. , A technique of transferring to a second computer and transferring the second computer to a second storage system connected to the second computer is disclosed.

  In US Pat. No. 6,209,002, a first storage system transfers data stored in a first storage system to a second storage system, and the second storage system further includes a third storage system. A technique for transferring data is disclosed. The computer and the first storage system are connected by a communication link, the first storage system and the second storage system are connected by a communication link, and the second storage system and the third storage system communicate with each other. Connected by a link. The first storage system holds the first logical volume to be replicated. The second storage system holds a second logical volume that is a replica of the first logical volume and a third logical volume that is a replica of the second logical volume. The third storage system holds a fourth logical volume that is a copy of the third logical volume. In this patent document, the second storage system exclusively performs data replication processing from the second logical volume to the third logical volume and data replication processing from the third logical volume to the fourth logical volume. To run.

US Pat. No. 5,170,480 US Pat. No. 6,209,002

  The technique disclosed in US Pat. No. 5,170,480 always uses a first computer and a second computer for data replication. The first computer performs normal business, and the load of data replication processing on the first computer cannot be ignored. Furthermore, since the data for duplication uses the communication link between the first computer and the first storage system, it collides with the data transfer required for normal business, and the data reference and data required for normal business There is a problem that the update time becomes long.

  The technique disclosed in US Pat. No. 6,209,002 requires a storage capacity that is twice the amount of data to be duplicated in the second storage system and the third storage system. In addition, since the amount of data to be replicated is large, the time spent for data replication processing is long, and the data in the third storage system becomes old. As a result, when the business is resumed using the data in the third storage system, there is a problem that the time until the data in the third storage system is updated becomes long and the time until the business is resumed is increased. Further, in this document, the first storage system stores data in the upper computer when the data marching process with the second storage system is completed in addition to the data update process in the first storage system. Report update completion. Therefore, the time spent for data update from the computer is long, and the longer the distance between the first storage system and the second storage system, the longer the time spent for data update. As a result, according to the technique disclosed in this document, there is a problem in that the distance between the storage systems cannot be made too large.

  An object of the present invention is to perform data transfer or data duplication between a plurality of storage systems without affecting the upper computer of the storage system. Furthermore, the communication between the storage system and the computer is not affected.

  Furthermore, an object of the present invention is to reduce the data storage area held in a plurality of storage systems. Furthermore, data transfer or data duplication is performed between the plurality of storage systems at high speed and efficiently so as not to affect the operations of the upper computers of the plurality of storage systems.

  In order to solve these problems, in the present invention, the first storage system stores information relating to the update of data stored in the first storage system as a journal. Specifically, the journal is composed of a copy of data used for the update and update information such as a write command at the time of update. Further, the second storage system acquires the journal via a communication line between the first storage system and the second storage system. The second storage system holds a copy of the data held by the first storage system, and corresponds to the data of the first storage system in the order of data update in the first storage system using the journal. Update the data.

  In the present invention, the second storage system holds a second storage area for storing a copy of data stored in the first storage area stored in the first storage system, Information related to data update is stored as a journal in a third storage area dedicated to the journal. The storage capacity of the third storage area can be smaller than that of the second storage area. Further, the third storage system acquires the journal via a communication line between the second storage system and the third storage system, and stores it in a fourth storage area dedicated to the journal. The storage capacity of the fourth storage area can be smaller than that of the second storage area. The third storage system holds a fifth storage area for storing a copy of the data stored in the second storage area, and uses the journal to update the data in the order of data update in the second storage area. Data in the fifth storage area corresponding to the second storage area is updated.

  According to the present invention, it is possible to provide a storage system capable of transferring data or copying data between a plurality of storage systems without affecting the upper computer of the storage system.

  Furthermore, it is possible to provide a storage system that does not affect the communication between the storage system and the computer.

  Furthermore, according to the present invention, the data storage area held in a plurality of storage systems can be reduced.

  Furthermore, it is possible to transfer data or copy data between a plurality of storage systems at high speed and efficiently so as not to affect the operations of a computer above the plurality of storage systems.

  An embodiment of a data processing system according to the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a block diagram showing a logical configuration of an embodiment of the present invention.

  In one embodiment of the present invention, the host computer 180 and the storage system 100A are connected by the connection path 190, and the storage system 100A and the storage system 100B that holds a copy of the data stored in the storage system 100A are connected by the connection path 200. It is a configuration. In the following description, in order to facilitate the distinction between the storage system 100 that holds the replication target data and the storage system 100 that holds the replication data, the storage system 100 that holds the replication target data is designated as a primary storage system 100A, The storage system 100 that holds the replicated data is called a secondary storage system 100B. The storage area of the storage system is divided and managed, and the divided storage area is referred to as a logical volume.

  The storage area of the storage system 100 is divided and managed, and the divided storage area is called a logical volume 230. The capacity of the logical volume 230 and the physical storage location (physical address) in the storage system 100 can be specified using a maintenance terminal such as a computer connected to the storage system 100 or the host computer 180. The physical address of each logical volume 230 is stored in volume information 400 described later. The physical address is, for example, a number (storage device number) that identifies the storage device 150 in the storage system 100 and a numerical value that uniquely indicates the storage area in the storage device, for example, a position from the beginning of the storage area of the storage device. . In the following description, the physical address is a set of a storage device number and a position from the beginning of the storage area of the storage device. In the following description, a logical volume is a storage area of one storage device, but it is also possible to associate one logical volume with storage areas of a plurality of storage devices by converting logical addresses and physical addresses.

  The reference and update of data stored in the storage system 100 is uniquely determined by a number (logical volume number) for identifying a logical volume and a numerical value uniquely indicating the storage area, for example, a position from the beginning of the storage area of the logical volume. Hereinafter, a set of a logical volume number and a position from the beginning of the storage area of the logical volume (position within the logical address) is referred to as a logical address.

  In the following description, in order to facilitate the distinction between the replication target data and the replication data, the replication target logical volume 230 is referred to as a primary logical volume, and the logical volume 230 that is replication data is referred to as a secondary logical volume. A pair of primary logical volume and secondary logical volume is called a pair. The relationship between the primary logical volume and the secondary logical volume, the status, and the like are stored in pair information 500 described later.

  In order to maintain the data update order between logical volumes, a management unit called a group is provided. For example, it is assumed that the host computer 180 updates data 1 of the primary logical volume 1, then reads data 1, and performs a process of updating the data 2 of the primary logical volume 2 using the numerical value of the data 1. When the data replication processing from the primary logical volume 1 to the secondary logical volume 1 and the data replication processing from the primary logical volume 2 to the secondary logical volume 2 are performed independently, from the replication processing of data 1 to the secondary logical volume 1 Before the data 2 may be copied to the secondary logical volume 2 in some cases. If the replication process of data 1 to the secondary logical volume 1 stops due to a failure or the like between the replication process of the data 2 to the secondary logical volume 2 and the replication process of the data 1 to the secondary logical volume 1, the secondary logical volume Data consistency between 1 and secondary logical volume 2 is lost. Even in such a case, in order to maintain the consistency of the data of the secondary logical volume 1 and the secondary logical volume 2, the logical volumes that need to follow the data update order are registered in the same group, and each time the data is updated, An update number of group information 600 (to be described later) is assigned, and replication processing is performed on the secondary logical volume in the order of the update numbers. For example, in FIG. 1, the logical volume (DATA1) and logical volume (DATA2) of the primary storage system 100A constitute group 1. A logical volume (COPY1) that is a secondary product of the logical volume (DATA1) and a logical volume (COPY2) that is a duplicate of the logical volume (DATA2) constitute a group 1 in the secondary storage system.

  When updating the data of the primary logical volume that is the data replication target, a journal, which will be described later, is created and stored in the logical volume in the primary storage system 100A in order to update the data of the secondary logical volume. In the description of this embodiment, a logical volume that stores only a journal (hereinafter referred to as a journal logical volume) is assigned to each group. In FIG. 1, a logical volume (JNL1) is assigned to group 1.

  A journal logical volume is also allocated to the group of the secondary storage system 100B. The journal logical volume is used for storing a journal transferred from the primary storage system 100A to the secondary storage system 100B. By storing the journal in the journal logical volume, for example, when the load on the secondary storage system 100B is high, the secondary logical volume data is not updated when the journal is received, and after a while, the load on the secondary storage system 100B is low. The data of the logical volume can also be updated. Further, when there are a plurality of connection lines 200, journal transfer from the primary storage system 100A to the secondary storage system 100B can be performed in a multiplexed manner, and the transfer capability of the connection line 200 can be used effectively. Due to the update order, a large number of journals may be accumulated in the secondary storage system 100B. However, a journal that cannot be used immediately for data update of the secondary logical volume may be released to the journal logical volume to release the cache memory. it can. In FIG. 1, a logical volume (JNL2) is assigned to group 1 in the secondary storage system.

  The journal is composed of write data and update information. The update information is information for managing write data, and stores the time when the write command is received, the group number, the update number of group information 600 described later, the logical address of the write command, the data size of the write data, and the write data. It consists of the logical address of the journal logical volume. The update information may hold only one of the time when the write command is received and the update number. When the creation time of the write command exists in the write command from the host computer 180, the creation time in the write command may be used instead of the time when the write command is received. An example of journal update information will be described with reference to FIGS. The update information 310 stores a write command received at 22:20:10 on March 17, 1999. The write command is a command for storing write data at a position 700 from the beginning of the storage area of logical volume number 1, and the data size is 300. Journal write data is stored from a position 1500 from the beginning of the storage area of logical volume number 4 (journal logical volume). It can be seen that the logical volume with logical volume number 1 belongs to group 1 and is the fourth data update from the start of group 1 data replication.

  For example, as shown in FIG. 3, the journal logical volume is divided into a storage area (update information area) for storing update information and a storage area (write data area) for storing write data. The update information area is stored in the order of the update number from the top of the update information area. When the end of the update information area is reached, the update information area is stored from the top of the update information area. The write data area stores write data from the beginning of the write data area, and stores from the beginning of the write data area when reaching the end of the write data area. The ratio between the update information area and the write data area may be a fixed value or may be set by the maintenance terminal or the host computer 180. These pieces of information are held in pointer information 700 described later. In the following explanation, the journal logical volume is divided into update information and write data areas and the journal logical volume is used. However, the journal, that is, the update information and write data are stored continuously from the beginning of the logical volume. May be adopted.

  The operation of reflecting the data update to the primary logical volume of the storage system 100A on the secondary logical volume of the secondary storage system 100B will be outlined with reference to FIG.

  (1) Upon receiving a write command for data in the primary logical volume from the host computer 180, the storage system 100A performs data update in the primary logical volume (DATA1) by command reception processing 210 and read / write processing 220 described later. The journal is stored in the journal logical volume (JNL1) (270 in FIG. 1).

  (2) The storage system 100B reads a journal from the storage system 100A by a journal read process 240 described later, and stores the journal in the journal logical volume (JNL2) by a read / write process 220 (280 in FIG. 1).

  (3) When the storage system 100A receives an instruction to read a journal from the storage system 100B, the storage system 100A reads the journal from the journal logical volume (JNL1) and transmits it to the storage system 100B by an instruction reception process 210 and a read / write process 220 described later. (280 in FIG. 1).

  (4) The storage system 100B reads the journal from the journal logical volume (JNL2) in ascending order of the update numbers using the pointer information 700 by the restore process 250 and the read / write process 220, which will be described later, and the secondary logical volume (COPY1). (290 in FIG. 1).

  The internal structure of the storage system 100 is shown in FIG. The storage system 100 includes one or more host adapters 110, one or more disk adapters 120, one or more cache memories 130, one or more shared memories 140, one or more storage devices 150, one or more common paths. 160 and one or more connecting lines 170. The host adapter 110, the disk adapter 120, the cache memory 130, and the shared memory 140 are connected to each other by a common path 160. The common path 160 may be duplicated when the common path 160 fails. The disk adapter 120 and the storage device 150 are connected by a connection line 170. Although not shown, a maintenance terminal for setting, monitoring, maintaining, etc. of the storage system 100 is connected to all the host adapters 110 and the disk adapters 120 using dedicated lines.

  The host adapter 110 controls data transfer between the host computer 180 and the cache memory 130. The host adapter 110 is connected to the host computer 180 or another storage system 100 through the connection line 190 and the connection line 200. The disk adapter 120 controls data transfer between the cache memory 130 and the storage device 150. The cache memory 130 is a memory that temporarily holds data received from the host computer 180 or data read from the storage device 150. The shared memory 140 is a memory shared by all the host adapters 110 and the disk adapters 120 in the storage system 100.

  The volume information 400 is information for managing the logical volume, and holds the volume status, format format, capacity, pair number, and physical address. FIG. 4 shows an example of the volume information 400. The volume information 400 is stored in a memory that can be referred to from the host adapter 110 and the disk adapter 120, for example, the management memory 140. The volume status holds one of “normal”, “normal”, “secondary”, “abnormal”, and “unused”. The logical volume 230 whose volume state is “normal” or “normal” indicates that the logical volume 230 is normally accessible from the host computer 180. The logical volume 230 whose volume state is “secondary” may permit access from the host computer 180. The logical volume 230 whose volume state is “primary” indicates that the logical volume 230 is being replicated. The logical volume 230 whose volume status is “secondary” indicates that the logical volume 230 is used for replication. The logical volume 230 whose volume state is “abnormal” indicates that the logical volume 230 cannot be normally accessed due to a failure. The failure is, for example, a failure of the storage device 150 that holds the logical volume 230. A logical volume 230 whose volume status is “unused” indicates that it is an unused logical volume 230. The pair number is effective when the volume state is “primary” or “secondary”, and holds a pair number for specifying pair information 500 described later. In the example shown in FIG. 4, the logical volume 1 stores data from the beginning of the storage area of the storage device 150 having the format format OPEN3, the capacity 3 GB, and the storage device number 1, and is accessible and is a replica of the data. Indicates that it is a subject.

  The pair information 500 is information for managing a pair, and holds a pair status, a primary storage system number, a primary logical volume number, a secondary storage system number, a secondary logical volume number, a group number, and a copied address. FIG. 5 shows an example of the pair information 500. The pair information 500 is stored in a memory that can be referred to from the host adapter 110 and the disk adapter 120, for example, the management memory 140. The pair status holds one of “normal”, “abnormal”, “unused”, “not copied”, and “copying”. When the pair status is “normal”, it indicates that data replication of the primary logical volume 230 is normally performed. When the pair status is “abnormal”, it indicates that the primary logical volume 230 cannot be replicated due to a failure. The failure is, for example, disconnection of the connection path 200. When the pair status is “unused”, it indicates that the information of the pair number is not valid. When the pair status is “copying”, it indicates that an initial copy process described later is being performed. When the pair status is “not copied”, it indicates that the initial copy processing described later has not been performed yet. The primary storage system number holds a number that identifies the primary storage system 100A that holds the primary logical volume 230. The secondary storage system number holds a number that identifies the secondary storage system 100B that holds the secondary logical volume 230. In the case of the primary storage system, the group number holds the group number to which the primary logical volume belongs. In the case of a secondary storage system, the group number to which the secondary logical volume belongs is held. The copied address will be described in the initial copy process described later. In the pair information 1 of FIG. 5, the data replication target is the primary logical volume 1 of the primary storage system 1, the data replication destination is the secondary logical volume 1 of the secondary storage system 2, and data replication processing is normally performed. It shows that.

  The group information 600 holds a group status, a pair set, a journal logical volume number, and an update number. FIG. 6 shows an example of the group information 600. The group information 600 is stored in a memory that can be referred to from the host adapter 110 and the disk adapter 120, for example, the management memory 140. The group status holds one of “normal”, “abnormal”, and “unused”. When the group state is “normal”, it indicates that at least one pair state of the pair set is “normal”. When the group state is “abnormal”, it indicates that all pair states of the pair set are “abnormal”. When the group status is “unused”, it indicates that the information of the group number is not valid. In the case of a primary storage system, the pair set holds the pair numbers of all the primary logical volumes belonging to the group indicated by the group number. In the case of a secondary storage system, the pair numbers of all secondary logical volumes belonging to the group indicated by the group number are held. The journal logical volume number indicates a journal logical volume number belonging to the group of the group number. The initial value of the update number is 1, and 1 is added when data is written to the primary logical volume in the group. The update number is stored in the journal update information and is used in the secondary storage system 100B to maintain the data update order. For example, the group information 1 in FIG. 6 includes pair information 1 and 2, primary logical volumes 1 and 2, and journal logical volume 4, and indicates that data duplication processing is normally performed.

  The pointer information 700 is information that is held for each group and manages the journal logical volume of the group. The update information area start address, the write data area start address, the update information latest address, the update information oldest address, and the write data latest Holds the address, the oldest address of the write data, the read start address, and the retry start address. 7 and 8 show an example of the pointer information 700. FIG. The update information area start address holds the start logical address of the storage area (update information area) for storing the update information of the journal logical volume. The write data area start address holds the start logical address of the storage area (write data area) for storing the write data of the journal logical volume. The latest update information address holds the first logical address used to store update information when the journal is stored next time. The update information oldest address holds the first logical address for storing the update information of the oldest journal (with the smallest update number). The latest write data address holds the first logical address used to store the write data when the journal is stored next time. The oldest write data address holds the first logical address for storing the write data of the oldest journal (with the smallest update number). The read start address and retry start address are used only in the primary storage system 100A, and are used in journal read reception processing described later. In the example of the pointer information 700 of FIGS. 7 and 8, the area for storing the management information of the journal (update information area) is from the beginning of the storage area of the logical volume 4 to the position 699, and the write data of the journal is stored. The area to be performed (write data area) is from the position 700 to the position 2699 in the storage area of the logical volume 4. The management information of the journal is stored from the position 200 to the position 499 of the storage area of the logical volume 4, and the management information of the next journal is stored from the position 500 of the storage area of the logical volume 4. The write data of the journal is stored from the position 1300 to the position 2199 of the storage area of the logical volume 4, and the write data of the next journal is stored from the position 2200 of the storage area of the logical volume 4.

  In the following description, one journal logical volume is assigned to one group, but a plurality of journal logical volumes may be assigned to one group. For example, two journal logical volumes are allocated to one group, pointer information 700 is provided for each journal logical volume, and journals are stored alternately. As a result, writing of the journal to the storage device 150 can be distributed, and an improvement in performance can be expected. Furthermore, the read performance of the journal is also improved. As another example, two journal logical volumes are assigned to one group, and usually only one journal logical volume is used. The other journal logical volume is used when the performance of the journal logical volume being used deteriorates. An example in which the performance decreases is when the journal logical volume is composed of a plurality of storage devices 150, holds data using the RAID 5 system, and one of the configured storage devices 150 is out of order.

  The volume information 400, the pair information 500, the group information 600, the pointer information 700, and the like described above are preferably stored in the shared memory 140. However, the present embodiment is not limited to this case, and these pieces of information may be stored or distributed in a concentrated manner in the cache memory 130, the host adapter 110, the disk adapter 120, and other storage devices 150.

  Next, a procedure for starting data replication from the primary storage system 100A to the secondary storage system 100B will be described with reference to FIGS.

  (1) Group creation will be described (step 900). Using the maintenance terminal or the host computer 180, the user refers to the group information 600 of the primary storage system 100A and acquires the group number A whose group status is “unused”. Using the maintenance terminal or the host computer 180, the user designates the group number A and gives a group creation instruction to the primary storage system 100A.

  Upon receiving the group creation instruction, the primary storage system 100A changes the group status of the designated group number A to “normal”.

  Similarly, the user refers to the group information 600 of the secondary storage system 100B, and acquires the group number B whose group status is “unused”. Using the maintenance terminal or the host computer 180, the user designates the secondary storage system 100B and the group number B, and issues a group creation instruction to the primary storage system 100A. The primary storage system 100A transfers the received group creation instruction to the secondary storage system 100B. The secondary storage system 100B changes the group status of the designated group number B to “normal”.

  The user may use the maintenance terminal of the secondary storage system 100B or the host computer 180 connected to the secondary storage system 100B to specify the group number B and give a group creation instruction to the secondary storage system 100B.

  (2) Pair registration will be described (step 910). Using the maintenance terminal or the host computer 180, the user designates information indicating the data replication target and information indicating the data replication destination, and issues a pair registration instruction to the primary storage system 100A. The information indicating the data replication target is the data replication target group number A and the primary logical volume number. The information indicating the data replication destination is the secondary storage system 100B that stores the replicated data, the group number B, and the secondary logical volume number.

  Upon receiving the pair registration instruction, the primary storage system 100A acquires the pair number whose pair information is “unused” from the pair information 500, sets the pair status to “not copied”, sets the primary storage system number to the primary storage system number, The primary storage system number indicating 100A is designated as the primary logical volume number, the designated primary logical volume number is designated as the secondary storage system number, the designated secondary storage system number is designated as the secondary logical volume number, and the designated secondary logical number is designated. The designated group number A is set in the volume number and the group number. The primary storage system 100A adds the acquired pair number to the pair set of the group information 600 of the designated group number A, and changes the volume state of the primary logical volume number to “primary”.

  The primary storage system 100A instructs the secondary storage system 100B of the primary storage system number indicating the primary storage system 100A, the group number B specified by the user, the primary logical volume number, and the secondary logical volume number. The secondary storage system 100B obtains an unused pair number from the pair information 500, sets the pair status to “not copied”, the primary storage system number, and the primary storage system number indicating the storage system 100A to the primary logical volume number. The designated primary logical volume number is designated as the secondary storage system number, the secondary storage system number indicating the secondary storage system B is designated as the secondary logical volume number, and the designated secondary logical volume number is designated as the group number. Group number B is set.

  The secondary storage system 100B adds the acquired pair number to the pair set of the group information 600 of the designated group number B, and changes the volume state of the secondary logical volume number to “secondary”.

  The above operation is performed on all data replication target pairs.

  In the above description, the processing of registering a logical volume in a group and setting a logical volume pair at the same time has been described, but may be performed individually.

  (3) Journal logical volume registration will be described (step 920). The user uses the maintenance terminal or the host computer 180 to issue an instruction (journal logical volume registration instruction) for registering a logical volume (journal logical volume) used to save a journal to the group to the primary storage system 100A. The journal logical volume registration instruction includes a group number and a logical volume number.

  The primary storage system 100A registers the designated logical volume number in the journal logical volume number of the group information 600 of the designated group number. The volume state of the volume information 400 of the logical volume is set to “normal”.

  Similarly, the user refers to the volume information 400 of the secondary storage system 100B using the maintenance terminal or the host computer 180, and designates the secondary storage system 100B, the group number B, and the logical volume number to be used as the journal logical volume. Then, journal logical volume registration is performed in the primary storage system 100A. The primary storage system 100A transfers the journal logical volume registration instruction to the secondary storage system 100B. The secondary storage system 100B registers the designated logical volume number in the journal logical volume number of the group information 600 of the designated group number B. The volume state of the volume information 400 of the logical volume is set to “normal”.

  Using the maintenance terminal of the secondary storage system 100B or the host computer 180 connected to the secondary storage system 100B, the user designates the group number, the logical volume number to be used as the journal logical volume, and issues a journal logical volume registration instruction to the secondary storage. You may go to system 100B.

  The above operation is performed for the logical volumes used as all journal logical volumes. The order of step 910 and step 920 is not the same.

  (4) The start of data replication processing will be described (step 930). Using the maintenance terminal or the host computer 180, the user designates the group number for starting the data replication process and instructs the primary storage system 100A to start the data replication process. The primary storage system 100A sets the copied addresses of all pair information 400 belonging to the designated group to 0.

  The primary storage system 100A instructs the secondary storage system 100B to start journal read processing and restore processing described later.

  The primary storage system 100A starts an initial copy process described later.

  (5) The end of the initial copy will be described (step 940).

  When the initial copy is completed, the primary storage system 100A notifies the secondary storage system 100B of the completion of the initial copy process. The secondary storage system 100B changes the pair status of all secondary logical volumes belonging to the designated group to “normal”.

  FIG. 10 is a flowchart of the initial copy process. The initial copy process creates a journal for each unit size in order from the beginning of the storage area, using the copied address of the pair information 500 for all storage areas of the primary logical volume to be replicated. The copied address has an initial value of 0, and the created data amount is added each time a journal is created. From the beginning of the storage area of the logical volume to the previous address before the copied address, the journal has been created by the initial copy process. By performing the initial copy process, it is possible to transfer data that has not been updated in the primary logical volume to the secondary logical volume. In the following description, it is described that the host adapter A in the primary storage system 100A performs processing, but the disk adapter 120 may perform the processing.

  (1) The host adapter A in the primary storage system 100A obtains a primary logical volume A whose pair status is “not copied” in a pair belonging to the group to be processed, and changes the pair status to “copying”. The following processing is repeated (steps 1010 and 1020). If the primary logical volume A does not exist, the process is terminated (step 1030).

  (2) If the logical volume A exists in step 1020, the host adapter A creates a journal for data of unit size (for example, 1 MB). The journal creation process will be described later (step 1040).

  (3) The host adapter A adds the data size of the created journal to the copied address (step 1050).

  (4) The above processing is repeated until the copied address reaches the capacity of the primary logical volume A (step 1060). When the copied address is equal to the capacity of the primary logical volume A, since the journal is created for all storage areas of the primary logical volume A, the pair status is updated to “normal” and the processing of the other primary logical volume is performed. Is started (step 1070).

  In the flowchart described above, the logical volumes are described as being targeted one by one, but a plurality of logical volumes may be processed simultaneously.

  FIG. 11 is a diagram for explaining the process of the command receiving process 210, FIG. 12 is a flowchart of the command receiving process 210, and FIG. 13 is a flowchart of the journal creating process. Hereinafter, the operation when the primary storage system 100A receives a write command from the host computer 180 to the logical volume 230 to be replicated will be described using these.

  (1) The host adapter A in the storage system 100A receives an access command from the host computer. The access command includes a read command, a write command, a journal read command to be described later, a logical address to be commanded, a data amount, and the like. Hereinafter, the logical address in the access instruction is assumed to be logical address A, the logical volume number is assumed to be logical volume A, the logical volume position is assumed to be logical volume position A, and the data amount is assumed to be data amount A (step 1200).

  (2) The host adapter A checks the access command (Steps 1210 and 1215). If it is determined in step 1215 that the access command is a journal read command, journal read reception processing described later is performed (step 1220). If the access command is other than the journal read command and write command, for example, if it is a read command, a read process is performed as in the conventional technique (step 1230).

  (3) If the access command is a write command in the check in step 1210, the volume status 400 is checked by referring to the volume information 400 of the logical volume A (step 1240). If it is determined in step 1240 that the logical volume A is in a state other than “normal” or “normal”, access to the logical volume A is impossible and an abnormal end is reported to the host computer 180 (step 1245). ).

  (4) If it is determined in step 1240 that the volume status of the logical volume A is “normal” or “normal”, the host adapter A secures the cache memory 130 and sends data to the host computer 180. Notify you that you are ready. The host computer 180 receives the notification and transmits the write data to the primary storage system 100A. The host adapter A receives the write data and stores it in the cache memory 130 (step 1250, 1100 in FIG. 11).

  (5) The host adapter A refers to the volume state of the logical volume A and checks whether the logical volume A is a data replication target (step 1260). If the volume status is “correct” in the examination in step 1260, the logical volume A is the data replication target, and journal creation processing described later is performed (step 1265).

  (6) If it is determined in step 1260 that the volume status is “normal”, or after completion of the journal creation processing in step 1265, the host adapter A writes the write data to the disk adapter 120 in the storage device 150. The command is issued (1140 in FIG. 11) and the completion is reported to the host computer 180 (steps 1270 and 1280). Thereafter, the disk adapter 120 stores the write data in the storage device 150 by the read / write process (1110 in FIG. 11).

  Next, journal creation processing will be described.

  (1) The host adapter A checks the volume state of the journal logical volume (step 1310). If it is determined in step 1310 that the volume status of the journal logical volume is “abnormal”, the journal status cannot be stored in the journal logical volume, so the group status is changed to “abnormal” and the process is terminated (step 1315). In this case, the journal logical volume is changed to a normal logical volume.

  (2) If it is determined in step 1310 that the journal logical volume is normal, the journal creation process is continued. The journal creation process differs depending on whether it is a process in the initial copy process or a command reception process (step 1320). If the journal creation process is a process within the command reception process, the process from step 1330 is performed. If the journal creation process is within the initial copy process, the process from step 1370 is performed.

(3) If the journal creation process is a process within the instruction reception process, the host adapter A checks whether the write target logical address A is the target of the initial copy process (step 1330). If the pair status of the logical volume A is “not copied”, journal creation processing is performed later in the initial copy processing, so the processing ends without creating a journal (step 1335). When the pair status of logical volume A is “Copying”, if the copied address is equal to or smaller than the position A in the logical address, journal creation processing will be performed later in the initial copy processing. The process is terminated (step 1335). Other than the above, that is, if the pair status of the logical volume A is “copying” and the copied address is greater than or equal to the position A in the logical address, or if the pair status of the logical volume A is “normal”, the initial copy processing has already been performed. Since it has been completed, continue the journal creation process.

  (4) Next, the host adapter A checks whether the journal can be stored in the journal logical volume. The pointer information 700 is used to check whether there is an unused area in the update information area (step 1340). If the latest update information address of the pointer information 700 is equal to the oldest address of the update information, the unused area does not exist in the update information area, and the process ends as a journal creation failure (step 1390).

  If it is determined in step 1340 that there is an unused area in the update information area, the pointer information 700 is used to check whether write data can be stored in the write data area (step 1345). If the sum of the write data latest address and the data amount A is equal to or larger than the write data oldest address, it cannot be stored in the write data area, and the process ends as journal creation failure (step 1390).

  (5) If the journal can be stored, the host adapter A acquires the update number, the logical address for storing the update information, and the logical address for storing the write data, and creates the update information in the cache memory 130. The update number is acquired from the group information 600 of the target group, and a value obtained by adding 1 is set as the update number of the group information 600. The logical address for storing the update information is the update information latest address of the pointer information 700, and a numerical value obtained by adding the size of the update information is set as the update information latest address of the pointer information 700. The logical address for storing the write data is the latest write data address of the pointer information 700, and a numerical value obtained by adding the data amount A to the write data latest address is set as the write data latest address of the pointer information 700.

  The host adapter A sets the obtained numerical value and group number, the time when the write command is received, the logical address A and the data amount A in the write command as update information (step 1350, 1120 in FIG. 11). For example, in the state of the group information 600 shown in FIG. 6 and the pointer information 700 shown in FIG. 7, when a write command having a data size of 100 is received at a position 800 from the beginning of the storage area of the primary logical volume 1 belonging to the group 1, Update information shown in FIG. 22 is created. The update number of the group information is 5, the update information update address of the pointer information is 600 (the update information size is 100), and the write data update address is 2300.

  (6) The host adapter A instructs the disk adapter 120 to write the journal update information and write data to the storage device 150 and terminates normally (step 1360, 1130, 1140, 1150 in FIG. 11).

  (7) If the journal creation process is a process within the initial copy process, the process from step 1370 is performed. The host adapter A checks whether a journal can be created. The pointer information 700 is used to check whether there is an unused area in the update information area (step 1370). If the latest update information address of the pointer information 700 is equal to the oldest address of the update information, the unused area does not exist in the update information area, and the process ends as a journal creation failure (step 1390). In the case of the initial copy processing shown in this embodiment, the write data of the journal is read from the primary logical volume and the write data area is not used, so it is not necessary to check the unused area of the write data area.

  (8) If it is determined in step 1370 that a journal can be created, the host adapter A acquires a numerical value set in the update information and creates the update information in the cache memory 130. The update number is acquired from the group information 600 of the target group, and a value obtained by adding 1 is set as the update number of the group information 600. The logical address for storing the update information is the position of the update information latest address of the pointer information 700, and a numerical value obtained by adding the size of the update information is set as the update information latest address of the pointer information 700.

  The host adapter A performs the initial copy to the acquired logical value and group number, the start time of this processing, the logical address of the initial copy processing target, the initial copy processing amount, and the logical address of the journal logical volume that stores the write data. A logical address to be processed is set (step 1380, 1120 in FIG. 11).

(9) The host adapter A instructs the disk adapter 120 to write the update information to the storage device 150 and terminates normally (step 1385, 1140, 1160 in FIG. 11).
).

  In the above description, the update information is described as existing in the cache memory 130, but may be stored in the shared memory 140 or the like.

  The writing of the write data to the storage device 150 does not have to be asynchronous, that is, not immediately after Step 1360 and Step 1385. However, when the host computer 180 issues a write command to the logical address A again, the write data in the journal is overwritten. Therefore, before the write data is received from the host computer 180, the write data in the journal It is necessary to write to the storage device 150 corresponding to the logical address of the journal logical volume. Alternatively, the information may be saved in another cache memory and later written in the storage device 150 corresponding to the logical address of the journal logical volume of the update information.

  In the above-described journal creation processing, the journal is stored in the storage device 150. However, a predetermined amount of cache memory 130 is prepared in advance for the journal, and after all the cache memory is used, the journal is stored in the storage device 150. May be saved. The amount of journal cache memory is specified from, for example, a maintenance terminal.

  The read / write process 220 is a process that is executed by the disk adapter 120 receiving a command from the host adapter 110 or the disk adapter 120. The processing to be executed is a process of writing data in the designated cache memory 130 to a storage area in the storage device 150 corresponding to the designated logical address, and a designation from the storage area in the storage device 150 corresponding to the designated logical address. For example, a process of reading data into the cache memory 130.

  FIG. 14 is a diagram for explaining the operation (journal read reception processing) of the host adapter A of the primary storage system 100A that has received the journal read command, and FIG. 15 is a flowchart. Hereinafter, the operation when the primary storage system 100A receives a journal read command from the secondary storage system 100B will be described using these.

  (1) The host adapter A in the primary storage system 100A receives an access command from the secondary storage system 100B. The access command includes an identifier indicating a journal read command, a command target group number, and the presence / absence of a retry command. Hereinafter, the group number in the access command is referred to as group number A (step 1220, 1410 in FIG. 14).

  (2) The host adapter A checks whether the group status of the group number A is “normal” (step 1510). If it is determined in step 1510 that the group status is other than “normal”, for example, “failure”, the secondary storage system 100B is notified of the group status and the processing is terminated. The secondary storage system 100B performs processing according to the received group state. For example, if the group status is “failure”, the journal read process is terminated (step 1515).

  (3) If the group status of group number A is “normal” in step 1510, the host adapter A checks the status of the journal logical volume (step 1520). If it is determined in step 1520 that the volume status of the journal logical volume is not “normal”, for example, if it is “failure”, the group status is changed to “failure”, and the secondary storage system 100B is notified of the group status, The process ends. The secondary storage system 100B performs processing according to the received group state. For example, if the group status is “failure”, the journal read process is terminated (step 1525).

  (4) If it is determined in step 1520 that the volume state of the journal logical volume is “normal”, it is checked whether the journal read instruction is a retry instruction (step 1530).

  (5) If the journal read command is a retry instruction in the examination in step 1530, the host adapter A transmits the previously transmitted journal to the secondary storage system 100B again. The host adapter A reserves the cache memory 130 and instructs the disk adapter to read the information on the size of the update information into the cache memory from the retry start address of the pointer information 700 (1420 in FIG. 14).

  In the read / write processing of the disk adapter, update information is read from the storage device 150, stored in the cache memory 130, and notified to the host adapter A (1430 in FIG. 14).

  The host adapter A receives a notification of update information read completion, acquires the logical address of the write data and the size of the write data from the update information, secures the cache memory 130, and reads the write data into the cache memory to the disk adapter. (Step 1540, 1440 in FIG. 14).

  In the read / write process of the disk adapter, the write data is read from the storage device 150, stored in the cache memory 130, and notified to the host adapter A (1450 in FIG. 14).

  The host adapter A receives the write data read completion notification, transmits update information and write data to the secondary storage system 100B, releases the cache memory 130 holding the journal, and ends the processing (step 1545, 14 of FIG. 14).

  (6) If it is determined in step 1530 that there is no retry instruction, the host adapter A checks whether there is a journal that has not been transmitted, and if it exists, transmits the journal to the secondary storage system 100B. The host adapter A compares the read start address of the pointer information 700 with the latest update information address (step 1550).

  If the read start address is equal to the latest update information address, all journals have been sent to the secondary storage system 100B, so "no journal" is sent to the secondary storage system 100B (step 1560), and the previous journal read command At this time, the storage area of the journal transmitted to the secondary storage system 100B is released (step 1590).

  In the journal storage area release processing, a retry start address is set in the update information oldest address of the pointer information 700. When the update information oldest address becomes the write data area start address, the update information oldest address is set to 0. The oldest write data address of the pointer information 700 is changed to a value obtained by adding the size of the write data transmitted in response to the previous read journal command. If the oldest write data address is a logical address that is greater than the capacity of the journal logical volume, the write data area start address is subtracted and corrected.

  (7) If there is an unsent journal in step 1550, the host adapter A secures the cache memory 130 and caches the size information of the update information from the read start address of the pointer information 700 in the disk adapter. A command to read into the memory is issued (1420 in FIG. 14).

  In the read / write process of the disk adapter A, the update information is read from the storage device 150, stored in the cache memory 130, and notified to the host adapter (1430 in FIG. 14).

  The host adapter A receives the notification of the completion of reading the update information, acquires the logical address of the write data and the size of the write data from the update information, secures the cache memory 130, and writes the write data to the disk adapter A in the cache memory. It is instructed to read (step 1570, 1440 in FIG. 14).

  In the read / write process of the disk adapter A, the write data is read from the storage device 150, stored in the cache memory 130, and notified to the host adapter (1450 in FIG. 14).

  The host adapter A receives the write data read completion notification, transmits the update information and the write data to the secondary storage system 100B (step 1580), and releases the cache memory 130 holding the journal (1460 in FIG. 14). ). Then, the read start address is set as the retry start address of the pointer information 700, and a numerical value obtained by adding the update information size of the transmitted journal to the read start address is set.

  (8) The host adapter A releases the storage area of the journal transmitted to the secondary storage system 100B when processing the previous journal read command (step 1590).

  In the journal read reception process described above, the primary storage system 100A transmits journals to the secondary storage system 100B one by one, but a plurality of journals may be simultaneously transmitted to the secondary storage system 100B. The number of journals to be transmitted with one journal read command may be specified in the journal read command by the secondary storage system 100B, or when the group is registered, etc., the user can store the journal number in the primary storage system 100A or the secondary storage system 100B. May be specified. Furthermore, the number of journals to be dynamically transmitted by one journal read command may be changed according to the transfer capability or load of the connection path 200 between the primary storage system 100A and the secondary storage system 100B. Also, the journal transfer amount may be specified in consideration of the write data size of the journal, not the number of journals.

  In the journal read reception process described above, the journal is read from the storage device 150 to the cache memory 130. However, if the journal exists in the cache memory 130, this process is not necessary.

  The journal storage area release process in the journal read reception process described above is performed at the time of the next journal read command process, but may be released immediately after the journal is transmitted to the secondary storage system 100B. Further, the secondary storage system 100B may set an update number that may be released in the journal read command, and the primary storage system 100A may release the journal storage area in accordance with the instruction.

  FIG. 16 is a diagram for explaining the journal read command processing 240, FIG. 17 is a flowchart, and FIG. 18 is a flowchart for journal storage processing. Hereinafter, an operation in which the host adapter B of the secondary storage system 100B reads the journal from the primary storage system 100A and stores it in the journal logical volume will be described using these.

  (1) The host adapter B in the secondary storage system 100B secures the cache memory 130 for storing the journal, an identifier indicating that it is a journal read command, the group number of the primary storage system 100A to be commanded, and the presence or absence of a retry instruction Is sent to the primary storage system 100A. Hereinafter, the group number in the access command is referred to as group number A (step 1700, 1610 in FIG. 16).

  (2) The host adapter B receives the response and journal of the primary storage system 100A (1620 in FIG. 16). The host adapter B checks the response, and if the response from the primary storage system 100A is “no journal”, the primary storage system 100A does not have a journal of the specified group, and after a certain time, the primary storage system 100A. The read journal command is transmitted to (steps 1720 and 1725).

  (4) If the response of the primary storage system 100A is “group status is failure” or “group status is not used”, the group status of the secondary storage system 100B is changed to the received status, and the journal read processing is terminated. (Steps 1730, 1735).

  (5) If the response of the primary storage system 100A is other than the above, that is, if the response ends normally, the volume status of the journal logical volume is examined (step 1740). When the volume status of the journal logical volume is “abnormal”, since the journal cannot be stored in the journal logical volume, the group status is changed to “abnormal” and the process is terminated (step 1745). In this case, the journal logical volume is changed to a normal logical volume and the group status is returned to normal.

  (6) If the volume status of the journal logical volume is “normal” in step 1740, journal storage processing 1800 described later is performed. When the journal storage process 1800 is completed normally, the next journal read command is transmitted. Alternatively, after a predetermined time elapses, the next journal read command is transmitted (step 1760). The next journal command may be transmitted at regular intervals at regular intervals. The number of received journals, the communication amount of the connection line 200, or the journal stored in the secondary storage system 100B is stored. It may be determined by the capacity, the load of the secondary storage system 100B, and the like. Furthermore, the storage capacity of the journal held by the primary storage system 100A or the pointer information of the primary storage system 100A may be read from the secondary storage system 100B and determined based on the numerical value. The information transfer may be performed with a dedicated command or may be included in a response to a journal read command. The subsequent processing is the same as that after step 1710.

  (7) If the journal storage process in step 1800 does not end normally, there is not enough unused area in the journal logical volume, so the received journal is discarded and a retry instruction journal read command is sent after a certain time (step) 1755). Alternatively, the journal is held in the cache memory, and journal storage processing is performed again after a predetermined time. This is because an unused area may increase in the journal logical volume after a certain period of time by performing a restore process 250 described later. In the case of this method, it is not necessary to give a retry instruction to the journal read instruction.

  Next, the journal storage process 1800 shown in FIG. 18 will be described.

  (1) The host adapter B checks whether the journal can be stored in the journal logical volume. The pointer information 700 is used to check whether there is an unused area in the update information area (step 1810). If the latest update information address of the pointer information 700 is equal to the oldest address of the update information, there is no unused area in the update information area, so the process ends as a journal creation failure (step 1820).

  (2) If there is an unused area in the update information area in the check in step 1810, it is checked whether write data can be stored in the write data area using the pointer information 700 (step 1830). If the sum of the write data latest address and the received write data amount of the journal is equal to or greater than the oldest write data address, the write data cannot be stored in the write data area, and the process ends as a journal creation failure ( Step 1820).

  (3) If the journal can be stored, the host adapter B changes the group number of the received update information and the logical address of the journal logical volume. The group number is changed to the group number of the secondary storage system 100B, and the logical address of the journal logical volume is changed to the latest write data address of the pointer information 700. The host adapter B changes the update information latest address of the pointer information 700 to a numerical value obtained by adding the update information size to the update information latest address. The host adapter B changes the write data latest address of the pointer information 700 to a numerical value obtained by adding the write data size to the write data latest address (step 1840).

  (4) The host adapter B instructs the disk adapter 120 to write update information and write data to the storage device 150, and terminates the processing as a successful journal creation (step 1850, 1630 in FIG. 16). Thereafter, the disk adapter 120 writes update information and write data to the storage device 150 by read / write processing, and releases the cache memory 130 (1640 in FIG. 16).

  In the journal storage processing described above, the journal is stored in the storage device 150. However, a predetermined amount of cache memory 130 is prepared for the journal, and all the cache memory is used before the journal is stored in the storage device 150. May be saved. The amount of journal cache memory is specified from, for example, a maintenance terminal.

  FIG. 19 is a diagram for explaining the restore processing 250, and FIG. 20 is a flowchart. Hereinafter, an operation in which the host adapter B of the secondary storage system 100B uses the journal to update data using these will be described. The restore processing 250 may be performed by the disk adapter 120 of the secondary storage system 100B.

  (1) The host adapter B checks whether the group status of the group number B is “normal” (step 2010). If it is determined in step 2010 that the group status is other than “normal”, for example, “failure”, the restoration process is terminated (step 2015).

  (2) If the group status is “normal” in step 2010, the volume status of the journal logical volume is checked (step 2020). If it is determined in step 2020 that the volume status of the journal logical volume is “abnormal”, the access is not possible, so the group status is changed to “abnormal” and the process is terminated (step 2025).

  (3) If the volume status of the journal logical volume is “normal” in step 2020, it is checked whether there is a journal to be restored. The host adapter B acquires the update information oldest address and the update information latest address of the pointer information 700. When the update information oldest address and the update information latest address are equal, there is no journal, so the restore process is temporarily terminated, and the restore process is resumed after a certain time (step 2030).

  (4) If it is determined in step 2030 that there is a journal to be restored, the following processing is performed on the journal having the oldest (minimum) update number. The update information of the journal having the oldest (minimum) update number is stored from the update information oldest address of the pointer information 700. The host adapter B reserves the cache memory 130 and instructs the disk adapter to read the information on the size of the update information from the oldest address of the update information into the cache memory 130 (1910 in FIG. 19).

  In the read / write processing of the disk adapter, update information is read from the storage device 150, stored in the cache memory 130, and notified to the host adapter B (1920 in FIG. 19).

  The host adapter B receives the notification of the completion of reading the update information, acquires the logical address of the write data and the size of the write data from the update information, secures the cache memory 130, and reads the write data into the cache memory to the disk adapter. (1930 in FIG. 19).

  In the read / write processing of the disk adapter, the write data is read from the storage device 150, stored in the cache memory 130, and notified to the host adapter (step 2040, 1940 in FIG. 19).

  (5) The host adapter B obtains the logical address of the secondary logical volume to be updated from the update information, and instructs the disk adapter to write the write data to the secondary logical volume (step 2050, 1950 in FIG. 19). In the read / write processing of the disk adapter, data is written to the storage device 150 corresponding to the logical address of the secondary logical volume, the cache memory 130 is released, and the host adapter is notified (1960 in FIG. 19).

  (6) The host adapter B receives the notification of the completion of the write processing of the disk adapter and releases the journal storage area. In the journal storage area release processing, the update information oldest address of the pointer information 700 is changed to a value obtained by adding the size of the update information. When the update information oldest address becomes the write data area start address, the write data area start address is set to 0. The oldest write data address of the pointer information 700 is changed to a value obtained by adding the size of the write data. If the oldest write data address is a logical address that is greater than the capacity of the journal logical volume, the write data area start address is subtracted and corrected. Thereafter, the host adapter B starts the next restoration process (step 2060).

  In the restore processing 250 described above, the journal is read from the storage device 150 to the cache memory 130. However, if the journal exists in the cache memory 130, this processing is unnecessary.

  In the journal read reception process and the journal read command process 240 described above, the journal to be transmitted by the primary storage system 100A is determined by the pointer information 700, but the journal to be transmitted by the secondary storage system 100B may be determined. For example, an update number is added to the journal read command. In this case, in order to obtain the logical address of the update information of the update number designated by the secondary storage system 100B in the journal read reception process, the logical information in which the update information is stored from the update number in the shared memory 140 of the primary storage system 100A. Provide a table or search method for obtaining addresses.

  In the journal read reception process and the journal read command process 240 described above, a journal read command is used, but a normal read command may be used. For example, the group information 600 and pointer information 700 of the primary storage system 100A are transferred to the secondary storage system 100B in advance, and the secondary storage system 100B reads the data (that is, journal) of the journal logical volume of the primary storage system 100A. .

  In the journal read reception process described above, it has been described that journals are transmitted from the primary storage system 100A to the secondary storage system 100B in the order of update numbers, but it is not necessary to transmit the journals in the order of update numbers. Further, a plurality of journal read commands may be transmitted from the primary storage system 100A to the secondary storage system 100B. In this case, in order to process the journals in the order of update numbers in the restore process, the secondary storage system 100B is provided with a table or search method for obtaining a logical address storing update information from the update numbers.

  In the data processing system of the present invention described above, the primary storage system acquires a journal, and the secondary storage system replicates data. Thus, the host computer connected to the primary storage system does not bear a load related to data replication. Further, by transferring the journal between the primary storage system and the secondary storage system, the communication line of the host computer connected to the primary storage system and the primary storage system is not used.

  FIG. 23 is a diagram illustrating a logical configuration of the second embodiment.

  The host computer 180 and the storage system 100C are connected by a connection path 190, the storage system 100C and the primary storage system 100A are connected by a connection path 200, and the primary storage system 100A and the secondary storage system 100B are connected by a connection path 200. . The storage system 100C updates the data of the logical volume (ORG1) and the data of the logical volume (DATA1) in the primary storage system 100A when updating data to the logical volume (ORG1) of the storage system 100C.

  As described in the first embodiment, when the primary storage system 100A updates the data to the primary logical volume (DATA1), the primary storage system 100A uses the command reception process 210 and the read / write process 220 described above to store the journal in the journal logical volume (JNL1). Saving is performed (2310).

  The secondary storage system 100B reads the journal from the primary storage system 100A by the journal read processing 240 described above, and stores the journal in the journal logical volume (JNL2) by the read / write processing 220 (2320).

  When the primary storage system 100A receives an instruction to read a journal from the secondary storage system 100B, the primary storage system 100A reads the journal from the journal logical volume (JNL1) by the instruction reception process 210 and the read / write process 220, and transmits it to the secondary storage system 100B ( 2320).

  The secondary storage system 100B reads the journal from the journal logical volume (JNL2) according to the update number by the restore process 250 and the read / write process 220 described above, and stores the secondary logical volume (COPY1) that is a copy of the primary logical volume (DATA1). The data is updated (2330). Thus, by updating data in the order of update numbers, it is possible to maintain data consistency between logical volumes.

  In the data processing system of the present invention described above, the primary storage system acquires a journal and stores it in a storage area dedicated to the journal. Further, the secondary storage system stores the journal received from the primary storage system in a journal dedicated storage area. The journal dedicated storage area can be made smaller than the data replication target storage area, and the data of the primary storage system can be copied to the secondary storage system with a smaller storage capacity.

  FIG. 24 is a diagram showing a logical configuration of the third embodiment.

  The host computer 180 and the storage system 100C are connected by a connection path 190, the storage system 100C and the primary storage system 100A are connected by a connection path 200, and the primary storage system 100A and the secondary storage system 100B are connected by a connection path 200. . As described in the related art, the storage system 100C updates the data of the logical volume (ORG1) and the data of the logical volume (DATA1) in the primary storage system 100A when updating data to the logical volume (ORG1) of the storage system 100C. I do.

  The primary storage system 100A makes the storage system 100C appear to have a primary logical volume (DATA1), but does not allocate an actual storage area, that is, the storage device 150. For example, a numerical value indicating that the storage device 150 is not assigned to the physical address of the volume information 400 is set. When the primary storage system 100A receives a write command to the data of the primary logical volume (DATA1) from the storage system 100C, the primary storage system 100A does not perform the processing of step 1270 in the command reception processing 210 described above, but stores the journal in the journal logical volume (JNL1). Only save is performed (2410).

  The secondary storage system 100B reads the journal from the primary storage system 100A by the journal read processing 240 described above, and stores the journal in the journal logical volume (JNL2) by the read / write processing 220 (2420).

  When the primary storage system 100A receives an instruction to read a journal from the secondary storage system 100B, the primary storage system 100A reads the journal from the journal logical volume (JNL1) by the instruction reception process 210 and the read / write process 220, and transmits it to the storage system 100B (2420). ).

  The secondary storage system 100B reads the journal from the journal logical volume (JNL2) according to the update number by the restore process 250 and the read / write process 220 described above, and the data of the secondary logical volume (COPY1) that is a copy of the logical volume (ORG1). Is updated (2430). Thus, by updating data in the order of update numbers, it is possible to maintain data consistency between logical volumes.

  In the data processing system of the present invention described above, when a failure occurs in the storage system 100C or the host computer 180 connected to the storage system 100C, the logical volume (COPY1) of the secondary storage system 100B is in the primary storage system 100A. By reflecting the journal (JNL1), the latest data can be referred to and updated by the host computer connected to the storage system 100B. Further, by storing only the journal without holding the data copy in the primary storage system 100A, it is possible to reduce the storage capacity required for data replication.

  Although the invention made by the present inventor has been specifically described based on the embodiments, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the invention. Needless to say.

It is a block diagram which shows the logical structure of one Embodiment of this invention. 1 is a block diagram of a storage system according to an embodiment of the present invention. It is a figure explaining the relationship between the update information and write data of one Embodiment of this invention. It is a figure explaining the example of the volume information of one Embodiment of this invention. It is a figure explaining the example of the pair information of one Embodiment of this invention. It is a figure explaining the example of the group information of one Embodiment of this invention. It is a figure explaining the example of the pointer information of one Embodiment of this invention. It is a figure explaining the structure of the journal logical volume of one Embodiment of this invention. It is a flowchart explaining the procedure which starts the replication of the data of one Embodiment of this invention. It is a flowchart explaining the initial copy process of one Embodiment of this invention. It is a figure explaining the command reception process of one Embodiment of this invention. It is a flowchart of the command reception process of one Embodiment of this invention. It is a flowchart of the journal creation process of one Embodiment of this invention. It is a figure explaining the journal read reception process of one Embodiment of this invention. It is a flowchart of the journal read reception process of one Embodiment of this invention. It is a figure explaining the journal read command process of one Embodiment of this invention. It is a flowchart of the journal read command process of one Embodiment of this invention. It is a flowchart of the journal storage process of one Embodiment of this invention. It is a figure explaining the restore process of one Embodiment of this invention. It is a flowchart of the restore process of one Embodiment of this invention. It is a figure explaining the example of the update information of one Embodiment of this invention. It is a figure explaining the example of the update information at the time of the journal creation process of one Embodiment of this invention. It is a figure which shows the 2nd Embodiment of this invention. It is a figure which shows the 3rd Embodiment of this invention.

Explanation of symbols

100 storage system 110 host adapter 120 disk adapter 130 cache memory 140 management memory 150 storage device 160 common path 170 connection line 180 between disk adapter and storage device 180 host computer

Claims (9)

A first storage system connected to a host device and transmitting / receiving data to / from the host device;
A second storage system connected to the first storage system and receiving data from the first storage system;
A third storage system connected to the second storage system and receiving data from the second storage system;
The first storage system has a first storage area in which data sent from the host device is written,
The second storage system has a second storage area in which a copy of data to be written to the first storage area and order information relating to the data are written.
The third storage system includes a third storage area for storing data read from the second storage area and order information related to the data, and an order stored in the third storage area. Having a fourth storage area in which data stored in the third storage area based on information is stored;
The remote copy system, wherein the data and order information stored in the third storage area are read from the second storage system at predetermined time intervals. The remote copy system according to claim 1,
The remote storage system according to claim 3, wherein the third storage system controls the predetermined time interval according to the amount of data read from the second storage area. The remote copy system according to claim 1, wherein
The third storage system is controlled to determine the predetermined time interval according to the amount of data transmitted and received between the second storage system and the third storage system. A remote copy system characterized by being. The remote copy system according to claim 1, wherein
The remote storage system according to claim 3, wherein the third storage system controls the predetermined time interval in accordance with a storage capacity held in the third storage area. The remote copy system according to claim 1,
The remote copy system, wherein the third storage system is controlled so as to determine the predetermined time interval according to a processing load of the third storage system itself. The remote copy system according to claim 1,
The third storage system reads information about the storage capacity held by the second storage area in the second storage system from the second storage system, and converts the information into the read information. Accordingly, the remote copy system is controlled so as to determine the predetermined time interval. The remote copy system according to claim 1,
The second storage system owns management information about the second storage area,
The third storage system reads management information about the second storage area owned by the second storage system from the second storage system, and according to the read management information The remote copy system is controlled so as to determine the predetermined time interval. The remote copy system according to claim 1,
The third storage system sends a command for requesting transmission of data stored in the second storage area and order information relating to the data to the second storage system at each predetermined time interval. Is to send,
The second storage system is configured to transmit data stored in the second storage area and order information regarding the data to the third storage system in response to the command. Remote copy system. A first storage system connected to a host device and transmitting / receiving data to / from the host device;
A second storage system connected to the first storage system and receiving data from the first storage system;
A third storage system connected to the second storage system and receiving data from the second storage system;
The first storage system has a first storage area in which data sent from the host device is written,
In the second storage system, a second storage area in which data received from the first storage system is written, data received from the first storage system, and order information relating to the data are written in third. Storage area, and
The third storage system includes a fourth storage area that stores data read from the third storage area and order information related to the data, and an order stored in the fourth storage area. Having a fifth storage area in which data is stored from the fourth storage area based on information,
The remote copy system, wherein data and order information stored in the fourth storage area are read from the second storage system at predetermined time intervals.

Images