JP2020149573A - Storage controller, storage system, storage control method, and storage control program - Google Patents

Abstract

To avoid a decrease in response performance due to a resynchronization process in replication.SOLUTION: A storage controller 40 comprises: a first memory unit 41 for storing update information 410 relating to replication from a first storage device 50 to a second storage device 51; a difference management unit 42 for managing difference information 420 that represents a difference in the data of the first and second storage devices; a detection unit 43 for detecting a prescribed event relating to the use state of the first memory unit 41; an acquisition unit 44 for acquiring transfer data 460 from the first storage device 50 when a prescribed event is detected and setting the store region of the transfer data 460 to an update exclusion state; a control unit 45 for releasing the update exclusion state relating to the transfer data 460 when storing the transfer data 460 transferred via a second memory unit 46 to the second memory unit 46, and controlling the difference management unit 42 so as to delete the difference information 420; and the second memory unit 46.SELECTED DRAWING: Figure 6

Description

The present invention relates to a technique for storing a copy of data stored in a storage device in another storage device.

In today's highly computerized society, various computer systems that build social infrastructure are in operation. Many of these computer systems have remote (standby, standby, secondary) sites separate from the main (operational, regular, primary) sites in order to achieve high availability and reliability. It is provided. Such a computer system builds a disaster recovery system that enables continuous provision of services at the remote site even if the main site is damaged by multiplexing and retaining the data at the main site at the remote site. ing.

Such a disaster recovery system executes replication, which is a process of transferring data between storage devices installed at a main site and a remote site, respectively. The data transfer method is roughly classified into the following three methods.

The first data transfer method is a synchronous copy method in which writing is performed to the logical disk on the remote site side in synchronization with writing to the logical disk on the main site side. In the synchronous copy method, after writing to both the main site side and the remote site side logical disks is completed, the completion report is sent to the host device that issued the write request. In the synchronous copy method, the write response to the host device is delayed by the communication time to the remote site.

The second data transfer method is an asynchronous copy method in which writing to the logical disk on the main site side and writing to the logical disk on the remote site side are performed asynchronously. In the asynchronous copy method, when the writing to the logical disk of the main site is completed, the completion report is sent to the host device that issued the write request. Then, in the asynchronous copy method, the updated contents for the logical disk of the main site are transmitted to the remote site at a predetermined timing thereafter. The asynchronous copy method improves the write response to the host device as compared with the synchronous copy method, but the data update order at the remote site does not always match the data update order at the main site. Therefore, in the asynchronous copy method, for example, if a disaster occurs during the execution of asynchronous copy, the integrity of the data at the remote site is not guaranteed.

The third data transfer method is a method in which the data at the remote site is updated in the order of updating the data at the main site (that is, a high-performance asynchronous copy method) in the above-mentioned asynchronous copy method. Hereafter, it will be referred to as the asynchronous order guarantee copy method. In the asynchronous order guarantee copy method, the logical disk of the main site is written, and an order guarantee buffer for storing the update data representing the update contents and the update sequence number is provided. Then, in the asynchronous order guarantee copy method, the update data stored in the order guarantee buffer is applied to the logical disk of the remote site while guaranteeing the order by the update sequence number. In the asynchronous copy method, the update data in the order guarantee buffer that has not yet been transferred from the main site to the remote site is lost at the time of the disaster, but the integrity of the data at the remote site is ensured.

In recent years, the level of demand for availability, reliability, performance, etc. of computer systems has been increasing year by year, and expectations for technology that enables efficient and reliable replication by the above-mentioned data transfer method have increased. There is.

As a technique related to such a technique, Patent Document 1 discloses a control device that controls a system that replicates data in a storage device of a replication source connected to a host computer to a storage device of a replication destination. ing. This device stores information including a copy pair of the copy source volume and the copy destination volume for which the write order is guaranteed, and the replication processing status in the copy pair. Then, when the usage of the buffer shared and used by a plurality of copy pairs at the time of data replication is equal to or more than the threshold value, this device replicates in at least one copy pair among the plurality of copy pairs sharing the buffer. Stop processing.

Further, Patent Document 2 discloses a storage system that matches the stored contents between volumes even when a plurality of remote copies are executed asynchronously. In this system, a plurality of primary storage control devices and a plurality of sub-memory control devices are connected by a plurality of routes, and asynchronous remote copying is performed between each first volume and each second volume. This system stores the write data transferred from the primary storage control device to the secondary storage control device in the write data storage unit. This system manages update order information including write time and sequential number. Then, this system updates the stored contents of each second volume by the updateable time point by determining the time point when each second volume can be updated based on the update order information.

Further, Patent Document 3 discloses a remote copy processing system between disk array devices that realizes asynchronous order-guaranteed copying using a buffer having a small storage capacity. This system includes a transmission data buffer that stores update data. In this system, when there is a write request for update data from the host to the disk array device, if the update data does not overlap with the last update data stored in the cache due to the previous write request, the update data To the cache. When the update data overlaps with the last update data, this system copies the last update data stored in the cache to the transmission data buffer and then writes the update data to the cache. In this way, this system writes data to the transmission data buffer only when there is a write in the same area.

Japanese Unexamined Patent Publication No. 2018-163521 JP-A-2007-264946 Japanese Patent No. 5170169

FIG. 8 is a block diagram showing a configuration of a general storage system 2 that copies data asynchronously with data update between the main site 60 and the remote site 70. That is, the storage system 2 is a system that executes replication by the above-mentioned asynchronous copy method (including the asynchronous order guarantee copy method). The main site 60 has a storage device 600 and a host device 630 that accesses the storage device 600. The remote site 70 has a storage device 700 and a host device 730 that accesses the storage device 700.

FIG. 9 is a sequence diagram showing a data copy and update operation by the storage system 2 illustrated in FIG. As illustrated in FIG. 9, the storage device 600 at the main site 60 selects the copy area 611 on the logical disk 610 when executing the asynchronous copy process at a predetermined timing. After setting the selected copy area 611 to the update exclusive (prohibited) state, the storage device 600 transmits the copy data of the copy area 611 to the storage device 700 at the remote site 70 via the communication network 80. The storage device 700 updates the copy area 711 in the logical disk 710 using the received copy data, and transmits the update completion report to the storage device 600 via the communication network 80. In the communication via the communication network 80, a certain communication delay time depending on the communication status occurs. The storage device 600 releases the update exclusive state of the copy area 611 by receiving the update completion report from the storage device 700.

Here, while the storage device 600 is performing the asynchronous copy process for the copy area 611 described above, the host device 630 issues an update command (write command) to the area of the logical disk 610 that overlaps with the copy area 611. And. In this case, since the update by the update command is waited until the update exclusive state of the copy area 611 is released, the update response time as illustrated in FIG. 9 occurs.

FIG. 10 is a diagram illustrating a transition of the amount of data stored in the order guarantee buffer 620 when a failure of the communication network 80 occurs in the storage system 2 that performs asynchronous order guarantee copy. In this case, as illustrated in FIG. 8, the storage device 600 includes an order guarantee buffer 620 that stores the update contents for the logical disk 610 and the update sequence number in association with each other.

In the example shown in FIG. 10, until time T ₀ a failure in a communication network 80 occurs, the storage device 600, so perform asynchronous order guarantee copying e.g. regularly for the storage device 700, it is stored in the order assurance buffer 620 The amount of data is maintained at a certain level. However, after the time T ₀ of a failure, by the process of transferring the update data stored in the order assurance buffer 620 from the storage device 600 to the storage device 700 stagnates, amount data stored in the order assurance buffer 620 Gradually increases. Then, the data amount reaches the upper limit value of the order assurance buffer 620 at time T ₁ (i.e., in the order assurance buffer 620, there is no newly stored a storage area to update data).

In this case, storage unit 600, the time T ₁ and later, since it is impossible to perform asynchronous order guarantee copy, after pausing the asynchronous order guarantee copying process is performed only in (difference management is performed only while the differential management Shift to mode). However, the difference management is to manage which area (part) has the difference between the logical disk 610 and the logical disk 710.

In the example shown in FIG. 10, after a failure in a communication network 80 is eliminated at time T _2, the storage device 600 executes the resynchronization process. However, the resynchronization process is to transfer (copy) data from the logical disk 610 to the logical disk 710 based on the result of the difference management, thereby transferring the data stored in the logical disk 610 to the logical disk 710. It is a process to match the stored data. The storage unit 600, by completing the re-synchronization processing at time T _3, after the data and matches stored in the data and the logical disk 710 stored in the logical disk 610, is paused Asynchronous order guarantee Resumes copy processing.

Here, consider the above-mentioned resynchronization process by the storage device 600. In the resynchronization process, when copying data in a certain area of the logical disk 610 to the logical disk 710, the storage device 600 needs to set the area to be copied to the update exclusive state in the same manner as the operation illustrated in FIG. is there. The reason is that if the data in a certain area is updated during copying in the resynchronization process, the consistency of the data cannot be maintained.

In this case, the host device 630 cannot perform the update process (write process) on the area to be copied until the update exclusive state is released. That is, for example, when the amount of data in the area to be copied is large, or when the communication delay time becomes long due to the occurrence of congestion in the communication network 80, the update response time as illustrated in FIG. 9 may be significantly deteriorated. In this case, the influence on the provision of services by the main site 60 is very large. Patent Documents 1 to 3 do not particularly mention such a problem. A main object of the present invention is to provide a storage control device or the like that solves this problem.

The storage control device according to one aspect of the present invention applies the update contents for the data stored in the first storage device to the duplication of the data stored in the second storage device. Difference representing the difference between the first storage means for storing the update information representing the contents, the data stored in the first storage device, and the duplication of the data stored in the second storage device. The first difference management means for managing information, a detection means for detecting a predetermined event related to the usage status of the first storage means, and the first method based on the difference information when the predetermined event is detected. The acquisition means for acquiring the transfer data to the second storage device and setting the storage area for the transfer data in the first storage device to the update exclusive state, and the second storage means. When the transfer data to be transferred to the second storage device via the second storage device is stored in the second storage means, the difference management is performed so as to release the update exclusion state and delete the difference information regarding the transfer data. A control means for controlling the means and the second storage means are provided.

From another viewpoint of achieving the above object, in the storage control method according to one aspect of the present invention, the information processing device stores the updated contents of the data stored in the first storage device in the second storage device. In the case of applying to the duplication of the data, the update information representing the update content is stored in the first storage means, and the data stored in the first storage device and the second storage device are stored. It manages the difference information representing the difference from the duplication of the data stored in the data, detects a predetermined event related to the usage status of the first storage means, and when the predetermined event is detected, the difference. Based on the information, the transfer data to the second storage device is acquired from the first storage device, and the transfer data storage area in the first storage device is set to the update exclusive state. When the transfer data to be transferred to the second storage device via the second storage means is stored in the second storage means, the update exclusive state is released and the difference information regarding the transfer data is deleted. To do.

Further, from the further viewpoint of achieving the above object, the storage control program according to one aspect of the present invention stores the updated contents for the data stored in the first storage device in the second storage device. In the case of applying to the duplication of the data, the data stored in the first storage device in a computer provided with the first storage means for storing the update information representing the update contents and the second storage means. A difference management process that manages difference information representing a difference from the duplication of the data stored in the second storage device, and a detection process that detects a predetermined event related to the usage status of the first storage means. When the predetermined event is detected, the transfer data to the second storage device is acquired from the first storage device based on the difference information, and the transfer data in the first storage device is acquired. When the acquisition process of setting the transfer data storage area to the update exclusive state and the transfer data to be transferred to the second storage device via the second storage means are stored in the second storage means. The update exclusive state is released, and a control process for controlling the difference management process so as to delete the difference information related to the transfer data is executed.

Further, the present invention can also be realized by a computer-readable, non-volatile recording medium in which the storage control program (computer program) is stored.

The present invention makes it possible to avoid a decrease in response performance when executing resynchronization processing between storage devices in a system in which replication between storage devices is performed asynchronously.

It is a block diagram which shows the structure of the storage system 1 which concerns on 1st Embodiment of this invention. It is a figure which illustrates the storage contents of the order guarantee buffer 110 which concerns on 1st Embodiment of this invention. Flow chart (1/2) showing an operation in which the storage device 100 according to the first embodiment of the present invention updates the data stored in the logical disk 181 or 182 based on the instruction issued from the host device 190. Is. Flow chart (2/2) showing an operation in which the storage device 100 according to the first embodiment of the present invention updates the data stored in the logical disk 181 or 182 based on the instruction issued from the host device 190. Is. FIG. 5 is a flowchart showing an operation in which the storage device 100 according to the first embodiment of the present invention executes an asynchronous order guarantee copy process. FIG. 5 is a flowchart showing an operation in which the storage device 100 according to the first embodiment of the present invention executes a resynchronization process. It is a block diagram which shows the structure of the storage control device 40 which concerns on 2nd Embodiment of this invention. It is a block diagram which shows the structure of the information processing apparatus 900 which can execute the storage control apparatus which concerns on each embodiment of this invention. The update of data between the main site 60 and the remote site 70 is a block diagram showing a configuration of a general storage system 2 that copies data asynchronously. It is a sequence diagram which shows the copy and update operation of data by a general storage system 2. FIG. 5 is a diagram illustrating a transition of the amount of data stored in the order guarantee buffer 620 when a failure of the communication network 80 occurs in the storage system 2 that performs asynchronous order guarantee copy.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

<First Embodiment>
FIG. 1 is a block diagram showing a configuration of a storage system 1 according to a first embodiment of the present invention. The storage system 1 is a disaster recovery system in which a main site 10 and a remote site 20 are connected so as to be able to communicate with each other via a communication network (communication line) 30.

The main site 10 is an operation system (regular system) site, and includes a storage device (storage control device) 100 and a host device 190. The host device 190 is an information processing device such as a server device. The storage device 100 stores data used by the host device 190. The main site 10 may include a plurality of host devices 190 that access the storage device 100.

The remote site 20 is a standby system (reserve system) site, and includes a storage device 200 and a host device 290 as in the main site 10.

The storage system 1 may include a plurality of remote sites 20. Further, each of the plurality of sites included in the storage system 1 may have both a function as a main site and a function as a remote.

The storage device 100 at the main site 10 replicates the data stored in the storage device 100 to the storage device 200 at the remote site 20. That is, the data stored in the storage device 200 is a duplicate of the data stored in the storage device 100.

The storage system 1 according to the present embodiment does not apply (reflect) the updated contents to the storage device 200 at the same timing as the update to the data stored in the storage device 100, and the data in the storage device 200. Will be updated in the order of updating the data in the storage device 100. That is, the storage system 1 according to the present embodiment employs the above-mentioned asynchronous order guarantee copy method as the replication method.

The storage device 100 according to the present embodiment has a function as a storage control device that controls the execution of replication by the asynchronous order guarantee copy method. The storage system 1 may include a storage control device as a device different from the storage device 100. Hereinafter, in the present embodiment, the configuration of the storage device 100 that operates as the storage control device at the main site 10 will be described.

The storage device 100 according to the present embodiment includes an order guarantee buffer (first and second storage units) 110, a difference management unit 120, a detection unit 130, an acquisition unit 140, a control unit 150, a logical disk access control unit 160, and a cache. It includes 170 and logical disks 181 and 182. Further, the storage device 200 at the remote site 20 also includes an order guarantee buffer 210, a cache 270, and logical disks 281 and 282, similarly to the storage device 100. The storage device 200 may also have a configuration corresponding to the difference management unit 120, the detection unit 130, the acquisition unit 140, the control unit 150, and the logical disk access control unit 160, similarly to the storage device 100. In the embodiment, the configuration of the storage device 200 will be omitted for convenience of explanation. Further, the storage device 200 may not include the order guarantee buffer 210.

The logical disks 181 and 182 in the storage device 100 are storage devices such as magnetic disks. The number of logical disks included in the storage device 100 is not limited to two. The storage device 100 may include one or more logical disks. The logical disks 181 and 182 store the data used by the host device 190 and the data to be replicated by the storage device 100.

The logical disk access control unit 160 controls access from the host device 190 to the logical disk 181 or 182. The logical disk access control unit 160 manages the logical disks 181 and 182 as disks belonging to the same order guarantee group. However, the order guarantee group represents a set of logical disks when data for which the update order needs to be guaranteed is stored across a plurality of logical disks.

The logical disk access control unit 160 manages the logical disk 181 and the logical disk 281 in the storage device 200 as disks having a pair relationship, and manages the logical disk 182 and the logical disk 282 in the storage device 200 as disks having a pair relationship. Manage as. That is, the storage device 100 according to the present embodiment replicates the data stored in the logical disk 181 to the logical disk 281 and replicates the data stored in the logical disk 182 to the logical disk 282. Do. The storage device 100 may replicate the data stored in the logical disks 181 or 182 to two or more logical disks, respectively.

When the logical disk access control unit 160 receives an update instruction (write instruction) for the data stored in the logical disk 181 or 182 from the host device 190, the logical disk access control unit 160 stores the write data included in the update instruction in the cache 170. To do. The cache 170 is a storage device such as a memory that temporarily stores the write data for the logical disk 181 or 182.

The logical disk access control unit 160 confirms whether or not the update area in the logical disk 181 or 182 indicated by the update instruction is set to the update exclusive state (update by another instruction or the like is prohibited). The logical disk access control unit 160 can update the update area on the logical disk 181 or 182 when the update exclusive state is not set by another instruction or the like for the update area. The other instructions described above include, for example, another update instruction for the update area, an instruction for performing replication regarding the update area, and the like.

When the update area is not set to the update exclusive state, the logical disk access control unit 160 sets the update area to the update exclusive state and uses the write data included in the update instruction stored in the cache 170. Then, the update for the update area is executed.

After the logical disk access control unit 160 updates the update area, the detection unit 130, which will be described later, provides a storage area in the sequence guarantee buffer 110 that can newly store the update data 111 representing the update content by the update instruction. The update data 111 is stored in the order guarantee buffer 110 when it is not notified that the insufficient event has occurred. The order guarantee buffer 110 shown in FIG. 1 is a storage device such as a memory or a magnetic disk, and may be a storage device in which the memory and the magnetic disk are mixed. The order guarantee buffer 110 stores update data 111-1, update data 111-2, and the like, which represent a plurality of update contents for the data stored in the logical disk 181 or 182.

FIG. 2 is a diagram illustrating the stored contents of the order guarantee buffer 110 according to the present embodiment. The order guarantee buffer 110 stores the items illustrated in FIG. 2 in association with each other. The update data identifier shown in FIG. 2 is an identifier that can identify the update data related to each update. The order guarantee number shown in FIG. 2 is information indicating the order in which the updates were performed, and in the order guarantee buffer 110 illustrated in FIG. 2, the update data 111-2 is next to the update represented by the update data 111-1. Indicates that an update has been made.

The update data 111-1 illustrated in FIG. 2 is updated for the area of the logical disk 181 of the storage device 100 whose address is "00000-00100", and the replication destination of the update data 111-1 is the storage device 200. It shows that it is the logical disk 281 of. Similarly, the update data 111-2 illustrated in FIG. 2 is updated for the area of the logical disk 182 of the storage device 100 whose address is "03000-05000", and the replication destination of the update data 111-2 is set. It represents the logical disk 282 of the storage device 200.

The difference management unit 120 shown in FIG. 1 is a difference bitmap representing the difference between the data stored in the logical disks 181 and 182 of the storage device 100 and the data stored in the logical disks 281 and 282 of the storage device 200. (Difference information) 121 is generated and managed. However, the difference bitmap 121 is information indicating the presence or absence of a difference between the two data with a predetermined particle size. The difference management unit 120 stores the generated difference bitmap 121 in a storage device such as a memory included in the storage device 100, for example.

The data (information) managed by the difference management unit 120 can specify the location where the difference between the data stored in the logical disks 181 and 182 and the data stored in the logical disks 281 and 282 exists. It may be information, and may be data in a format different from that of the difference bitmap 121.

When the logical disk access control unit 160 updates the data to be updated on the logical disk 181 or 182, the difference management unit 120 generates a difference bitmap 121 regarding the data to be updated. The difference management unit 120 also deletes the difference bitmap 121 related to the data to be updated when the logical disk access control unit 160 stores the update data 111 related to the data to be updated in the order guarantee buffer 110. When the update data 111 related to the data to be updated is stored in the order guarantee buffer 110, the update data 111 is not yet reflected in the duplication of the data stored in the logical disk 281 or 282. However, when the difference management unit 120 determines that the update data 111 is reflected in the data duplication to be updated by storing the update data 111 in the order guarantee buffer 110, the difference bitmap regarding the update data 111 Delete 121.

The detection unit 130 shown in FIG. 1 detects a predetermined event related to the usage status of the order guarantee buffer 110 by referring to the order guarantee buffer 110. More specifically, the detection unit 130 confirms in the order guarantee buffer 110 whether or not an event has occurred in which the storage area that can newly store the update data 111 representing the update content by the update instruction has become insufficient. That is, in the present embodiment, the above-mentioned predetermined event means that "the order guarantee buffer 110 lacks a storage area that can newly store the update data 111". Then, the detection unit 130 notifies the logical disk access control unit 160 and the acquisition unit 140, which will be described later, of the detection result.

The event that the storage area that can newly store the update data 111 becomes insufficient due to the increase in the usage rate of the order guarantee buffer 110 causes, for example, a failure in the communication network 30 shown in FIG. 1 as described above with respect to FIG. It is caused by doing. That is, due to the occurrence of the failure, the process of transmitting the update data 111 stored in the order guarantee buffer 110 from the storage device 100 to the storage device 200 is delayed, so that the update data 111 stored in the order guarantee buffer 110 is delayed. The amount of data will gradually increase. Then, since the amount of data reaches the upper limit value of the order guarantee buffer 110, the order guarantee buffer 110 cannot newly store the update data 111.

The logical disk access control unit 160 executes an asynchronous order guarantee copy process from the logical disk 181 or 182 to the logical disk 281 or 282 in the background, for example, at a predetermined timing (for example, periodically).

When the update data 111 exists in the order guarantee buffer 110, the logical disk access control unit 160 transmits the update data 111 stored in the order guarantee buffer 110 to the storage device 200. The storage device 200 stores the update data 111 received from the storage device 100 in the cache 270 via the order guarantee buffer 210, and then writes the update data 111 stored in the cache 270 to the logical disk 281 or 282. As a result, the storage device 200 updates the copy of the data stored in the logical disk 181 or 182, which is stored in the logical disk 281 or 282.

The storage device 200 transmits to the storage device 100 that the update indicated by the update data 111 has been completed. When the logical disk access control unit 160 in the storage device 100 receives from the storage device 200 that the update indicated by the update data 111 is completed, the logical disk access control unit 160 deletes the update data 111 from the order guarantee buffer 110.

As described above, the logical disk access control unit 160 also performs an asynchronous order guarantee copy when the detection unit 130 detects an event in which the storage area that can newly store the update data 111 representing the update content by the update instruction is insufficient. Decides to suspend replication by. Then, after the replication by the asynchronous order guarantee copy is stopped, the storage device 100 shifts to the mode in which only the difference management is performed with respect to the logical disks 181 and 182 and the logical disks 281 and 282 by the difference bitmap 121. To determine. This mode corresponds to a mode in which only the difference management at time T ₁ to T ₂ shown in FIG. 10 described above.

After the storage device 100 shifts to the mode of performing the difference management by the difference bitmap 121, the difference management unit 120 generates the difference bitmap 121 regarding the update data 111 stored in the order guarantee buffer 110. After that, the control unit 150, which will be described later, deletes the update data 111 stored in the order guarantee buffer 110.

After the storage device 100 shifts to the mode in which only the difference management is performed by the difference bitmap 121, the acquisition unit 140 and the control unit 150 store the data stored in the logical disks 181 and 182 and the data stored in the logical disks 281 and 282. Executes the resynchronization process to match the data. Resynchronization process is a process executed in time T _2, to T ₃ shown in FIG. 10 described above. For example, as illustrated in FIG. 10, the acquisition unit 140 and the control unit 150 may perform the resynchronization process after the failure in the communication network 30 is resolved.

The acquisition unit 140 determines the area of data in the logical disk 181 or 182 to be transferred (transmitted) to the storage device 200 based on the difference bitmap 121. When the storage area capable of storing the determined data area exists in the order guarantee buffer 110, the acquisition unit 140 transfers the data (transfer data) from the determined data area on the logical disk 181 or 182 to the storage device 200. ) Is read.

The control unit 150 stores the transfer data read by the acquisition unit 140 in the order guarantee buffer 110. When the storage device 100 shifts to the mode in which only the difference management is performed by the difference bitmap 121, the control unit 150 deletes the update data 111 stored in the order guarantee buffer 110, so that the acquisition unit 140 and the control unit 140 and the control unit 150 delete the update data 111. At the time when the resynchronization process is started by the unit 150, the order guarantee buffer 110 can store the transfer data.

The acquisition unit 140 sets the data area in the logical disk 181 or 182 to the update exclusive state at the timing when the data area is determined as described above. The control unit 150 controls the difference management unit 120 so as to delete the difference bitmap 121 related to the data area at the timing when the transferred data is stored in the order guarantee buffer 110, and sets the update exclusive state regarding the data area. To release. Then, the control unit 150 executes the resynchronization process by transmitting the transfer data stored in the order guarantee buffer 110 to the storage device 200.

When the transfer data is stored in the order guarantee buffer 110 by the acquisition unit 140 with respect to the data area indicated by the difference bitmap 121, the difference management unit 120 deletes the difference bitmap 121 under the control of the control unit 150. To do. Then, the control unit 150 completes the resynchronization process by transferring data to the storage device 200 for all the difference bitmaps 121 existing in the difference management unit 120.

The storage device 100, at the timing when completing the re-synchronization processing (time T ₃ illustrated in FIG. 10 described _above) is resumed to perform a replication with asynchronous order guarantee copy that has been paused.

Next, the operation (processing) of the storage device (storage control device) 100 according to the present embodiment will be described in detail with reference to the flowcharts of FIGS. 3A and 3B, FIGS. 4 and 5.

3A and 3B are flowcharts showing an operation in which the storage device 100 according to the present embodiment updates the data stored in the logical disk 181 or 182 based on the instruction issued from the host device 190.

The logical disk access control unit 160 receives an update command for the data stored in the logical disk 181 or 182 from the host device 190 (step S101). The logical disk access control unit 160 stores the write data included in the received update instruction in the cache 170 (step S102). The logical disk access control unit 160 confirms whether or not the update exclusive state is set to the update exclusive state by another instruction or the like with respect to the update area indicated by the update instruction (step S103).

When the update area is set to the update exclusive state (Yes in step S104), the process returns to step S103. When the update area is not set to the update exclusive state (No in step S104), the logical disk access control unit 160 sets the update area to the update exclusive state (step S105).

The logical disk access control unit 160 updates the data to be updated on the logical disk 181 or 182 by using the write data included in the update instruction stored in the cache 170 (step S106). The difference management unit 120 generates a difference bitmap 121 regarding the data to be updated (step S107).

The detection unit 130 confirms in the order guarantee buffer 110 whether or not an event has occurred in which the storage area that can newly store the update data 111 indicating the update content by the update instruction is insufficient (step S108). When the event that the storage area capable of storing the update data 111 is insufficient has not occurred (No in step S109), the logical disk access control unit 160 stores the update data 111 in the order guarantee buffer 110 (step S110). ..

The difference management unit 120 deletes the difference bitmap 121 related to the data to be updated (step S111). The logical disk access control unit 160 releases the update exclusive state of the update area (step S112), and the entire process ends.

When an event occurs in which the storage area capable of storing the update data 111 is insufficient (Yes in step S109), the logical disk access control unit 160 releases the update exclusive state of the update area (step S113). The logical disk access control unit 160 suspends replication by the asynchronous order guarantee copy (step S114).

The difference management unit 120 generates a difference bitmap 121 regarding the update data 111 stored in the order guarantee buffer 110 (step S115). The control unit 150 deletes the update data 111 stored in the order guarantee buffer 110 (step S116), and the entire process ends.

FIG. 4 is a flowchart showing an operation in which the storage device 100 according to the present embodiment executes the asynchronous order guarantee copy process.

The logical disk access control unit 160 confirms whether or not the update data 111 exists in the order guarantee buffer 110 (step S201). If the update data 111 does not exist in the order guarantee buffer 110 (No in step S202), the process returns to step S201. When the update data 111 exists in the sequence guarantee buffer 110 (Yes in step S202), the logical disk access control unit 160 transmits the update data 111 stored in the sequence guarantee buffer 110 to the storage device 200 (step S203). ).

The logical disk access control unit 160 receives from the storage device 200 that the update indicated by the update data 111 for the logical disk 281 or 282 has been completed (step S204). The logical disk access control unit 160 deletes the update data 111 transmitted to the storage device 200 from the order guarantee buffer 110 (step S205), and the process returns to step S201.

FIG. 5 is a flowchart showing an operation in which the storage device 100 according to the present embodiment executes the resynchronization process.

The difference management unit 120 confirms whether or not the difference bitmap 121 exists (step S301). If the difference bitmap 121 does not exist (No in step S302), the entire process ends. When the difference bitmap 121 exists (Yes in step S302), the acquisition unit 140 determines the area of data on the logical disk 181 or 182 that transfers data to the storage device 200 based on the difference bitmap 121 (step S302). S303).

The acquisition unit 140 confirms whether or not a storage area capable of storing the determined data area exists in the order guarantee buffer 110 (step S304). If there is no area in the order guarantee buffer 110 that can store the determined data area (No in step S305), the process returns to step S304. When an area capable of storing the determined data area exists in the order guarantee buffer 110 (Yes in step S305), the acquisition unit 140 sets the determined data area in the update exclusive state (step S306).

The acquisition unit 140 reads the data area determined from the logical disk 181 or 182, and the control unit 150 stores the read transfer data in the order guarantee buffer 110 (step S307). The control unit 150 deletes the difference bitmap 121 regarding the determined data area by controlling the difference management unit 120 (step S308). The control unit 150 releases the update exclusion state of the determined data area (step S309). The control unit 150 transmits the transfer data stored in the order guarantee buffer 110 to the storage device 200, deletes the transfer data from the order guarantee buffer 110 (step S310), and returns to step S301.

The storage device 100 (storage control device) according to the present embodiment can avoid a decrease in response performance when executing resynchronization processing between storage devices in a system that asynchronously replicates between storage devices. it can. The reason is that when the storage device 100 executes data transfer associated with the resynchronization process between the storage device 100 and the storage device 200 via the sequence guarantee buffer 110 and stores the transferred data in the sequence guarantee buffer 110. This is because the update exclusion state related to the transferred data is released and the difference bitmap 121 related to the transferred data is deleted.

The effects realized by the storage device 100 according to the present embodiment will be described in detail below.

For example, as shown in FIGS. 8 to 10, in a general disaster recovery system (storage system 2), when a failure occurs in the communication network 80 connecting the main site 60 and the remote site 70, the failure is resolved and then the failure is resolved. A resynchronization process is performed to match the data stored in the logical disk 610 with the data stored in the logical disk 710. In this resynchronization process, the storage device 600 at the main site 60 puts the copied area in the update exclusive state in order to guarantee the integrity of the data when copying the data of a certain area on the logical disk 610 to the logical disk 710. Must be set. In this case, the host device 630 cannot perform the update process (write process) on the area to be copied until the update exclusive state is released. That is, for example, when the amount of data in the area to be copied is large, or when the communication delay time becomes long due to the occurrence of congestion in the communication network 80, the update response time illustrated in FIG. 9 may be significantly deteriorated. is there.

In response to such a problem, the storage device (storage control device) 100 according to the present embodiment includes an order guarantee buffer (first and second storage units) 110, a difference management unit 120, and a detection unit 130. It includes an acquisition unit 140 and a control unit 150, and operates as described above with reference to, for example, FIGS. 1 to 5. That is, the order guarantee buffer 110 applies the update contents for the data stored in the storage device (first storage device) 100 to the duplication of the data stored in the storage device (second storage device) 200. In this case, the update data 111 representing the update content is stored. The difference management unit 120 manages a difference bitmap (difference information) 121 representing the difference between the data stored in the storage device 100 and the duplicate of the data stored in the storage device 200. The detection unit 130 detects a predetermined event related to the usage status of the order guarantee buffer 110. When the predetermined event is detected, the acquisition unit 140 acquires the transfer data from the storage device 100 to the storage device 200 based on the difference bitmap 121, and the storage area of the transfer data in the storage device 100. Is set to the update exclusive state. Then, when the control unit 150 stores the transfer data to be transferred to the storage device 200 via the order guarantee buffer 110 in the order guarantee buffer 110, the control unit 150 releases the update exclusion state and the difference bitmap regarding the transfer data. The difference management unit 120 is controlled so as to delete 121.

That is, the storage device 100 according to the present embodiment transfers data via the sequence guarantee buffer 110 in the resynchronization process, and stores the transfer data in the storage device 100 at the timing when the transfer data is stored in the sequence guarantee buffer 110. The update exclusive state of the area is released, and the difference bitmap 121 related to the transferred data is deleted. Therefore, the storage device 100 stores the transferred data in the order guarantee buffer 110, and then manages the content of the update generated for the storage area of the transferred data as a new difference bitmap 121, so that the data can be stored. It is possible to release the update exclusive state of the transfer data storage area at an early timing while maintaining the consistency. As a result, the storage device 100 according to the present embodiment can shorten the period for setting the storage area of the data to be transferred in the resynchronization process to the update exclusive state, so that when the resynchronization process is executed between the storage devices. It is possible to avoid a decrease in response performance.

Further, the difference management unit 120 according to the present embodiment deletes the difference bitmap 121 related to the update data 111 when the update data 111 is stored in the order guarantee buffer 110, and then the detection unit 130 detects a predetermined event. In this case, the difference bitmap 121 regarding the update data 111 stored in the order guarantee buffer 110 is generated. Then, when a predetermined event is detected by the detection unit 130, the control unit 150 deletes the update data 111 stored in the order guarantee buffer 110. That is, the difference management unit 120 according to the present embodiment manages the minimum necessary difference bitmap 121, and the control unit 150 manages the update data stored in the order guarantee buffer 110 before performing the resynchronization process. 111 is deleted promptly. As a result, the storage device 100 according to the present embodiment can efficiently perform the resynchronization process via the order guarantee buffer 110 described above.

Further, in the storage device 100 according to the above-described embodiment, the order guarantee buffer 110 (first storage unit) for storing the update data 111 is used as a buffer (second storage unit) for storing the transfer data in the resynchronization process. Used as a part). That is, the storage device 100 has a configuration in which the cost is reduced by sharing the first storage unit and the second storage unit, but the storage device 100 uses a dedicated buffer as the second storage unit. You may be prepared.

In this case, the storage device 100 registers the update data 111 by the update command issued from the host device 190 in the first storage unit and the transfer data by executing the resynchronization process in the second storage unit. Can be done in parallel. In this case, the storage device 100 may perform common order management for the update data 111 and the transfer data. Further, when the storage device 100 includes a dedicated buffer as the second storage unit, the order guarantee buffer 110 is used to secure an area for storing the transfer data before the resynchronization process is executed. It is not necessary to delete the update data 111 from the buffer 110.

Further, the control unit 150 according to the present embodiment may have a function of receiving information indicating the status of the communication network 30 that communicably connects the storage device 100 and the storage device 200 from the outside. In this case, the control unit 150 transfers data in the resynchronization process when the information indicating the status of the communication network 30 and the amount of data to be transferred from the storage device 100 to the storage device 200 satisfy the conditions. To execute.

More specifically, for example, when the amount of data to be transferred is large, the control unit 150 suspends the data transfer in the resynchronization process until the congestion in the communication network 30 is greatly improved. Then, for example, when the amount of data to be transferred is small, the control unit 150 may start data transfer in the resynchronization process when the congestion in the communication network 30 is improved to some extent. The storage device 100 according to the present embodiment can perform the resynchronization process more efficiently by providing the control unit 150 that operates in this way.

Further, the replication method adopted by the storage system 1 according to the present embodiment is not limited to the asynchronous order guarantee copy method. The function of the storage device 100 according to the present embodiment as a storage control device is a system that includes a buffer for executing replication asynchronously with data update at the main site, for example, a system that executes asynchronous copy without guaranteeing the order. It can also be used in such cases.

<Second embodiment>
FIG. 6 is a block diagram showing the configuration of the storage control device 40 according to the second embodiment of the present invention.

The storage control device 40 according to the present embodiment includes a first storage unit 41, a difference management unit 42, a detection unit 43, an acquisition unit 44, a control unit 45, and a second storage unit 46.

When the first storage unit 41 applies the update contents for the data 500 stored in the first storage device 50 to the copy 510 of the data 500 stored in the second storage device 51, the update is applied. The update information 410 indicating the content is stored.

The difference management unit 42 manages the difference information 420 representing the difference between the data 500 stored in the first storage device 50 and the duplicate 510 of the data 500 stored in the second storage device 51.

The detection unit 43 detects a predetermined event 430 regarding the usage status of the first storage unit 41.

When the predetermined event 430 is detected, the acquisition unit 44 acquires the transfer data 460 from the first storage device 50 to the second storage device 51 based on the difference information 420, and also acquires the transfer data 460 to the second storage device 51 and the first storage. The storage area of the transfer data 460 in the device 50 is set to the update exclusive state.

When the control unit 45 stores the transfer data 460 to be transferred to the second storage device 51 via the second storage unit 46 in the second storage unit 46, the control unit 45 releases the update exclusive state and transfers the transfer data. The difference management unit 42 is controlled so as to delete the difference information 420 regarding the 460.

The storage control device 40 according to the present embodiment can avoid a decrease in response performance when executing resynchronization processing between storage devices in a system in which replication between storage devices is performed asynchronously. The reason is that the storage control device 40 executes data transfer associated with the resynchronization process between the first storage device 50 and the second storage device 51 via the second storage unit 46, and transfers data. This is because when the 460 is stored in the second storage unit 46, the update exclusive state regarding the transfer data 460 is released and the difference information 420 regarding the transfer data 460 is deleted.

<Hardware configuration example>
In each of the above-described embodiments, each part of the storage control device (storage device) shown in FIGS. 1 and 6 can be realized by a dedicated HW (HardWare) (electronic circuit). Further, in FIGS. 1 and 6, at least the following configuration can be regarded as a function (processing) unit (software module) of the software program.
A storage control function in the order guarantee buffer 110, the first storage unit 41, and the second storage unit 46.
・ Difference management units 120 and 42,
・ Detection units 130 and 43,
・ Acquisition units 140 and 44,
・ Control units 150 and 45,
-Logical disk access control unit 160.

However, the division of each part shown in these drawings is a configuration for convenience of explanation, and various configurations can be assumed at the time of mounting. An example of the hardware environment in this case will be described with reference to FIG.

FIG. 7 is a diagram illustrating an example of a configuration of an information processing device 900 (computer) capable of executing a storage control device according to each embodiment of the present invention. That is, FIG. 7 shows the configuration of a computer (information processing device) capable of realizing the storage control device shown in FIGS. 1 and 6, and provides a hardware environment capable of realizing each function in the above-described embodiment. Represent.

The information processing apparatus 900 shown in FIG. 7 includes the following as components.
-CPU (Central_Processing_Unit) 901,
-ROM (Read_Only_Memory) 902,
・ RAM (Random_Access_Memory) 903,
-Hard disk (storage device) 904,
-Communication interface 905,
・ Bus 906 (communication line),
A reader / writer 908 that can read and write data stored in a recording medium 907 such as a CD-ROM (Compact_Disc_Read_Only_Memory),
-Input / output interface 909 for monitors, speakers, keyboards, etc.

That is, the information processing device 900 including the above components is a general computer in which these components are connected via the bus 906. The information processing apparatus 900 may include a plurality of CPUs 901 or may include a CPU 901 configured by a multi-core processor.

Then, the present invention described by taking the above-described embodiment as an example supplies the computer program capable of realizing the following functions to the information processing apparatus 900 shown in FIG. 7. The function is the above-described configuration in the block configuration diagram (FIGS. 1 and 6) referred to in the description of the embodiment, or the function of the flowchart (FIGS. 3A and 3B, FIGS. 4 and 5). The present invention is then achieved by reading, interpreting, and executing the computer program in the CPU 901 of the hardware. Further, the computer program supplied in the device may be stored in a readable / writable volatile memory (RAM 903) or a non-volatile storage device such as a ROM 902 or a hard disk 904.

Further, in the above case, as a method of supplying the computer program into the hardware, a general procedure can be adopted at present. As the procedure, for example, there are a method of installing in the device via various recording media 907 such as a CD-ROM, a method of downloading from the outside via a communication line such as the Internet, and the like. Then, in such a case, the present invention can be regarded as being composed of the code constituting the computer program or the recording medium 907 in which the code is stored.

The invention of the present application has been described above using the above-described embodiment as a model example. However, the present invention is not limited to the above-described embodiments. That is, the present invention can apply various aspects that can be understood by those skilled in the art within the scope of the present invention.

1 Storage system 10 Main site 100 Storage device 110 Order guarantee buffer 111 Update data 120 Difference management unit 121 Difference bitmap 130 Detection unit 140 Acquisition unit 150 Control unit 160 Logical disk access control unit 170 Cache 181 and 182 Logical disk 190 Host device 20 Remote Site 200 Storage device 210 Order guarantee buffer 270 Cache 281 and 282 Logical disk 30 Communication network 40 Storage control device 41 First storage unit 410 Update information 42 Difference management unit 420 Difference information 43 Detection unit 430 Predetermined event 44 Acquisition unit 45 Control Part 46 Second storage 460 Transfer data 50 First storage device 500 Data 51 Second storage device 510 Duplicate 2 Storage system 60 Main site 600 Storage device 610 Logical disk 611 Copy area 620 Buffer 70 Remote site 700 Storage device 710 Logic Disk 711 Copy area 80 Communication network 900 Information processing device 901 CPU
902 ROM
903 RAM
904 hard disk (storage device)
905 Communication interface 906 Bus 907 Recording medium 908 Reader / writer 909 Input / output interface

Claims (10)

When the update content for the data stored in the first storage device is applied to the duplication of the data stored in the second storage device, the first storage for storing the update information representing the update content. Means and
A difference management means for managing difference information representing a difference between the data stored in the first storage device and the duplication of the data stored in the second storage device.
A detection means for detecting a predetermined event related to the usage status of the first storage means, and
When the predetermined event is detected, the transfer data to the second storage device is acquired from the first storage device based on the difference information, and the transfer data in the first storage device is acquired. And the acquisition method to set the storage area of
When the transfer data to be transferred to the second storage device via the second storage means is stored in the second storage means, the update exclusive state is released and the difference information regarding the transfer data is released. A control means that controls the difference management means to be deleted, and
With the second storage means
Storage control unit. The first storage means is provided as the second storage means.
The storage control device according to claim 1. When the update information is stored in the first storage means, the difference management means deletes the difference information related to the update information, and then the detection means detects the predetermined event. Generate the difference information regarding the update information stored in the first storage means,
When the predetermined event is detected by the detection means, the control means deletes the update information stored in the first storage means.
The storage control device according to claim 2. The difference management means manages the difference information that can identify the place where the difference exists in the data stored in the first storage device.
The storage control device according to any one of claims 1 to 3. As the predetermined event, the detection means detects that the first storage means lacks a storage area that can newly store the update information.
The storage control device according to any one of claims 1 to 4. The control means receives from the outside information representing the status of a communication network that communicably connects the first storage device and the second storage device, and the information representing the status of the communication network and the transfer. When the data amount of the data satisfies the condition, the transfer data is transferred from the first storage device to the second storage device.
The storage control device according to any one of claims 1 to 5. The first storage means is associated with the order in which the updates represented by the update information occur, and the update contents are applied to the duplication of the data stored in the second storage device according to the order. Memorize information,
The storage control device according to any one of claims 1 to 6. The storage control device according to any one of claims 1 to 7.
With the first storage device
With the second storage device
Storage system with. Depending on the information processing device
When the update content for the data stored in the first storage device is applied to the duplication of the data stored in the second storage device, the update information representing the update content is used as the first storage means. Remember,
The difference information representing the difference between the data stored in the first storage device and the duplication of the data stored in the second storage device is managed.
Detecting a predetermined event related to the usage status of the first storage means,
When the predetermined event is detected, the transfer data to the second storage device is acquired from the first storage device based on the difference information, and the transfer data in the first storage device is acquired. Set the storage area of
When the transfer data to be transferred to the second storage device via the second storage means is stored in the second storage means, the update exclusive state is released and the difference information regarding the transfer data is released. delete,
Storage control method. When the update content for the data stored in the first storage device is applied to the duplication of the data stored in the second storage device, the first storage for storing the update information representing the update content. On a computer with means and a second storage means,
A difference management process for managing difference information representing a difference between the data stored in the first storage device and the duplication of the data stored in the second storage device.
A detection process for detecting a predetermined event related to the usage status of the first storage means, and
When the predetermined event is detected, the transfer data to the second storage device is acquired from the first storage device based on the difference information, and the transfer data in the first storage device is acquired. Acquisition process to set the storage area of
When the transfer data to be transferred to the second storage device via the second storage means is stored in the second storage means, the update exclusive state is released and the difference information regarding the transfer data is released. A control process that controls the difference management process so that it is deleted,
Storage control program to execute.

Images