JP2003108315A - Storage subsystem - Google Patents

Abstract

PROBLEM TO BE SOLVED: To permit data migration between old and new systems without shutdown. SOLUTION: A plurality of first access routes 20, 21 from a CPU 10 to an old CU 13 (old subsystem) having an old VOL 14 under its control are prepared, and, between the old CU 13 and a new CU 11 (new subsystem) having a new VOL 12 under its control, a plurality of third access routes 30, 31 are set. Then, by splitting over more than once, connections are changed from the first access routes 20, 21 of the old subsystem of transmission source to the second access routes 20', 21' of the new subsystem of transmission destination. During the connection change, if the new subsystem receives an access from the CPU 10 through the second access routes 20', 21' of the new subsystem side, a path migration control part 111 relays the access, through the third access routes 30, 31, to the old subsystem side, where the access is processed. The data migration from the old subsystem to new one is executed after all the first access routes are switched to the second access routes.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a subsystem migration technique, and more particularly to a technique effectively applied to a migration operation of a subsystem under the control of a central processing unit in an information processing system or the like on the premise of non-stop operation. Regarding

[0002]

2. Description of the Related Art Data migration is the migration of data in an old disk subsystem, which functions as an external storage device in an information processing system, into a new disk subsystem. Conventionally, generally, a method of migrating data between disk subsystems is to stop access from the central processing unit to the device to be migrated, the CPU reads the data from the old disk subsystem, and the CPU reads the data from the new disk subsystem. A method involving CPU intervention, such as writing to a system, is known. According to this method, the customer's business for the disk subsystem is stopped for a long time during data migration.

On the other hand, HODM (Hitach) of Hitachi, Ltd. has been proposed as a technique for shortening business downtime by allowing access from the CPU even during data migration.
iOnline Data Migration function, IBM's extended remote copy function (XR
C) or peer remote copy function (hereinafter PPR
C) (Literature example IBM 3990 Model 6 E
and the Symm of EMC.
etrix Data Migration Serv
ice (SDMS) (reference example SYMMETRIX I
CDA Family PRODUCT ANNOUN
CEMENT SUMMMARY November 6
th, 1995).

In the HODM system, first, the CPU stops access to the old disk subsystem. After that, the connection path from the CPU to the old disk subsystem is changed from the CPU to the new disk subsystem, and a new access path is provided between the new and old disk subsystems. After that, the data of the old disk subsystem is read from the new disk subsystem through the new access path to start the migration and restart the access from the CPU. Both the old and new disk subsystems process the CPU access to the migrated area. In addition, CP for the unfinished migration area
At the time of U access, the data read from the old disk subsystem is reflected in the new disk subsystem and processed. This enables data migration during access from the CPU. A major feature of this function is that the old disk subsystem does not need to have a data migration function.

In the system based on XRC, the old disk subsystem has a function to reserve the write data from the CPU in the disk controller, and the CPU has a function to read the secured data. By writing this data to the new disk subsystem, it is possible to transfer data during access from the CPU. This method is characterized in that after the migration is completed, the customer's work for the old disk subsystem is stopped and the new disk subsystem is switched to.

In the PPRC system, the old disk subsystem and the new disk subsystem are connected to each other, and both have communication functions. By writing the write data of the CPU to the old disk subsystem to the new disk subsystem via this communication function, it is possible to transfer the data during access from the CPU. In addition, this system has a feature that, like XRC, the business of the customer is stopped and switched after the migration is completed.

On the other hand, the SDMS first stops the access from the CPU to the old disk subsystem. After that, the connection path from the CPU to the old disk subsystem is changed from the CPU to the new disk subsystem, and a new access path is provided between the new and old disk subsystems. After that, the migration is started by reading the data of the old disk subsystem from the new disk subsystem through the new access path. Further, after the shift is started, the access from the CPU is restarted. When the CPU access is to the migrated area, it is directly processed by the new disk subsystem. When accessing the migration incomplete area, the data of the track is read from the old disk subsystem, and then the normal processing is performed in the new disk subsystem. This enables data migration during access from the CPU.

[0008]

In the prior art as described above, the access to the data stored in the old disk subsystem can be stopped by making the CPU accessible even during data migration. The time to switch from the system to the new disk subsystem has been reduced. However, in the case of system control data such as an OS, even a temporary access suspension is a suspension of customer business, and the migration work has a large effect. In particular, there is a technical problem that the increasing number of customers who need online business for 24 hours cannot allow this, and the system can be moved only when the system is stopped such as the year-end and New Year holidays.

Further, in general, one subsystem can be used by being connected to a plurality of CPUs. At this time, in the subsystem, for each access route received,
Alternatively, the CPU is divided and processed in units of access route groups. Equivalent access to the other subsystem must also be partitioned.

In addition, CP is continued while CPU access is continued.
When the access route from U is switched to the new subsystem, the CPU recognizes that it is continuing to access the same device. After the data migration is completed and the old subsystem is removed, the CPU may issue a device information input request for device confirmation or the like. In the CPU that is checking the device and access route based on the match / mismatch between the device information read in the past and the device information read this time, if the information of the new subsystem is sent at this time, the CPU does not match the device information and the access route There is a concern that the access route will be cut and the subsystem will be down because it is judged that there is a failure.

An object of the present invention is to provide a subsystem migration technique capable of continuing access from a higher-level device to the subsystem side even during switching from the old subsystem to the new subsystem.

Another object of the present invention is to provide a subsystem migration technique capable of performing data migration in an uninterrupted state without requiring any suspension of access from a higher-level device to the subsystem side in the data migration procedure. Especially.

Another object of the present invention is to provide a subsystem migration technique capable of smoothly migrating an old subsystem operating under the control of a plurality of host devices to a new subsystem under non-stop operation. To provide.

Another object of the present invention is to avoid the occurrence of a failure due to an environmental change such as device information accompanying the transition from the old subsystem to the new subsystem, and to enable smooth subsystem migration. It is to provide subsystem migration technology.

[0015]

Generally, a plurality of access paths are provided from a host device such as a CPU or a channel to a subordinate subsystem, and the host device freely selects and switches the access path to access the subsystem. I do.
For example, an access route different from the original access route may be selected and used even at the time of resumption after an interruption that occurs in a series of instructions issued by a certain input / output processing request. Since the instructions before and after the interruption are a series of processing, it goes without saying that the subsystem cannot execute the instruction after the interruption unless the instruction before the interruption is executed.
Therefore, even if the access route is changed, the subsystem can recognize it and process it as a series of instructions.

According to the present invention, a plurality of first access paths from the host device to the old subsystem and a third access path between the new and old subsystems are provided.
When the access route is provided, the first access route from the host device to the old subsystem is divided into a plurality of times to change the connection to the second access route between the host device and the new subsystem. While the connection is being changed, the first and second access routes are connected to both the old and new subsystems from the host device. At this time, the old and new subsystems are equivalent if access is received from the host device. Access to the partner subsystem via the third access path, so that the access request is relayed.
By doing so, the partner subsystem executes the instruction before the interruption, so that the instruction after the interruption can be executed. Also, this equivalent access needs to be performed by both the old and new subsystems, but if the subsystem to be processed at the center during connection change is determined, it may be performed by the opposite subsystem. Also, when the processing request from the higher-level device is not interrupted, or when the processing request from the higher-level device comes with a fixed access route, etc., when the partner subsystem does not receive the next consecutive command from the higher-level device. May not need to make equivalent access to the other subsystem in some cases. Also, by making the access to the partner subsystem fixed to the access route to the partner subsystem, the partner subsystem is prevented from receiving the next consecutive command from the host device, and is received from the host device. You can also access it differently. In this way, the access route is divided into multiple times without stopping access from the host device,
It is possible to switch the connection from the old subsystem to the new subsystem.

For example, in the data migration in the disk subsystem, the old disk subsystem is the main processing center during the connection change, and the new disk subsystem passes through the third access path of the present invention. If the access request is relayed, the connection can be switched without stopping the access from the host device. However, if switching is performed while data migration is being performed, the old disk subsystem may be directly accessed by the higher-level device during switching, and data may be updated only in the old disk subsystem. If the data of the part that has already been migrated is updated, the data of that part will be missed.

Therefore, according to the present invention, by relaying an access request from the host device via the second access path to the old subsystem via the third access path, the new disk subsystem is transferred from the old disk subsystem. When the connection switching to the new disk subsystem is realized, the data transfer from the old subsystem to the new subsystem is started after the connection switching of all the first access paths to the second access paths is completed. , Prevents updates that do not go through the new disk subsystem in the part after data migration, and does not require data migration again. On the other hand, conversely, when the new disk subsystem is made to mainly perform the processing during the connection change, that is, during the connection change from the first access path to the second access path, the first access path is used. The old subsystem relays the access request from the higher-level device to the new subsystem via the third access path, so that the connection can be switched without stopping the access from the higher-level device. However, before or during data migration, when data that has not been migrated to the new disk subsystem is accessed by the host device, it cannot be processed.

Therefore, according to the present invention, in this case, prior to switching the first access path to the old subsystem to the second access path to the new subsystem, the old subsystem is switched to the new subsystem. After the data migration (copying) is completed in advance, the old disk subsystem is made to perform the operation of relaying the access request coming from the host device through the first access path to the new subsystem through the third access path. Thus, the connection switching from the old disk subsystem to the new disk subsystem is realized.

Further, in order to make it possible to switch the connection during the data transfer from the old subsystem to the new subsystem, if the data of both the old and new disk subsystems are constantly updated during the switching. Good.

Therefore, in the present invention, in such a case, in both the new and old disk subsystems,
By relaying access requests coming from the host device via the first and second access paths to other subsystems via the third access path, connection switching can be performed during data migration. .

Further, according to the present invention, in order to distinguish the access of a plurality of host devices and notify the other subsystem, at least the same number of third access paths as the number of host devices connected to the old disk subsystem are used. It is prepared for between systems. The access via the respective third access paths between the disk subsystems is made to correspond to the access from each higher-level device, and the data migration is possible when the old disk subsystem is connected to a plurality of higher-level devices. In addition, the number of third access routes includes not only the number of physical access routes but also the number of logical access routes.

Further, in the present invention, the new disk subsystem issues a device information input request to the old disk subsystem in advance, and at this time, the device information returned from the old disk subsystem is read and stored. . Then, in response to a device information input request from the host device, the stored device information of the old disk subsystem is sent instead of the device information of the new disk subsystem.

[0024]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described in detail below with reference to the drawings.

(Embodiment 1) FIG. 1 is a conceptual diagram showing an example of the configuration and operation of a general-purpose computer system, which is an embodiment of an information processing system in which the subsystem migration method of the present invention is implemented. .

The configuration of the present embodiment includes a CPU 10 which is a central processing unit, a new disk controller unit 11 (hereinafter referred to as a new CU 11) which is a data migration destination, and a new disk volume 12 (hereinafter referred to as a new VOL 12). Old subsystem, the old disk controller unit 13 (hereinafter referred to as the old CU13) and the old disk volume 14 (hereinafter called the old VOL14) that are the data migration source
It is composed of an old subsystem consisting of.

The old VOL 14 is a storage medium that operates under the control of the old CU 13, and stores data exchanged with the CPU 10 via the old CU 13.

The new VOL 12 is a storage medium that operates under the control of the new CU 11, and stores data exchanged with the CPU 10 via the new CU 11 and data transferred from the old VOL 14.

Further, the new CU 11 uses the second access route 2
0 ', the access request coming from the CPU 10 via the second access route 21', the third access route 30, the third access route
By relaying to the old CU 13 via the access route 31, as will be described later, the path migration control unit 111 that enables path migration between the new and old subsystems without stopping CPU access, and the data migration are controlled. The data migration control unit 112 is provided. Path transfer control unit 111
Performs an operation equivalent to that of the CPU 10 when accessing the old CU 13 using the third access route 30 and the third access route 31.

In the data migration processing of this embodiment, the CP
A first access route 20 and a first access route 2 which are connected between U10 and the old CU 13 which is originally used.
1 is the second access route 20 'and the second access route 21' between the CPU 10 and the new CU 11 which is the data migration destination.
The connection is switched as follows, and the new third access route 30 and the third access route 3 are provided between the new CU 11 and the old CU 13.
1 is used for connection. Also, new CU11 and new VOL
12 and between the old CU 13 and the old VOL 14 are respectively connected via a device path 12a and a device path 14a.

First, an example of data migration processing in the information processing system having the configuration illustrated in FIG. 1 will be described with reference to the flowchart of FIG. FIG. 6 shows operations 5a, 5b, ... Of maintenance personnel who perform migration of subsystem paths according to the first invention in the embodiment of FIG. ． ． ． ． , 5i and the CPU access path 50 indicating the access path used by the CPU 10 that changes accordingly, and the path migration control unit 11
Path migration control unit designation 5 indicating the state designated as 1
1. The processing CU 52 indicating the CU that processes the access from the CPU 10 is shown in time series.

First, since the normal processing is in progress before the maintenance staff starts the operation, the first access paths 20 and 21 are used from the CPU 10, and the CPU access path 50 indicates the first access paths 20 and 21. Further, since the path transfer control unit designation 51 is not related to the path transfer, when it is accessed by the CPU 10, it is processed by the new CU 11, and the same access to the old CU 13 through the third access paths 30 and 31 (which will be provided later) is performed. It is designated not to do so (hereinafter, such designation is referred to as processing by the own CU). Further, the processing CU 52 is the old CU 13. First, the maintenance staff uses the third access route 30 between the new CU 11 and the old CU 13.
And 31 are newly installed (operation 5a).

Further, the path transfer control unit 111 is provided with a CPU.
When the access is received from 10, the new CU 11 does not process the old CU1 through the third access routes 30 and 31.
The same access to 3 is performed (the access is relayed) (hereinafter, such a designation is referred to as processing by another CU) (operation 5b).

Next, from the CPU 10 to the first access route 2
0 is set to the offline state (operation 5c), and the access of the CPU 10 using the route is stopped.

At this time, the CPU access path 50 is only the first access path 21. Then, the connection is switched from the original configuration to the first access route 20 to the second access route 20 '(operation 5d).

When the connection is completed, the second access route 20 '(original first access route 20) is brought online from the CPU 10 (operation 5e).

As shown in the CPU access route 50, the CPU 10 starts the access using the first access route 21 and the access using the second access route 20 'as the access to the old CU 13. To be done. Although the new CU 11 receives the access using the second access route 20 ′, the path migration control unit 11
The third access route 3 without processing at the new CU 11 at 1
Make the same access to the old CU13 via 0 and 31,
It is operated so as to be processed by the old CU 13. Therefore C
From the PU, the old CU 13 is accessed by using either the second access route 20 '(first access route 21),
The processing only by the old CU 13 is continued. In the same procedure, the first access route 21
To the second access route 21 '. In this way, the access route from the CPU 10 to the old CU 13 can be switched to the new CU 11 without stopping the access. Finally, since all the switching has been completed, if the designation of the path migration control unit 111 is changed to the processing in the own CU (operation 5f), the new CU 11 changes to C as shown in the processing CU52.
The access request from the PU 10 is processed.
In this way, the processing subsystem can be switched from the old CU 13 and old VOL 14 to the new CU 11 and new VOL 12 without stopping access from the CPU 10.

Here, the timing of starting the data migration in the subsystem migration accompanied by the data migration as in the present embodiment illustrated in FIG. 1 will be examined. The data migration process can be performed if only one of the third access routes 30 and 31 is used. However, if the first access route 20 is switched to the new CU 11 as the second access route 20 'and before the first access route 21 is switched to the second access route 21', data migration is performed (started). If so, there is a possibility that the data update from the CPU 10 to the old CU 13 may be performed to the already-migrated portion through the first access route 21. If such an update were to be made, the new CU 11 would not know this, which would result in data migration failure.

Therefore, in the present embodiment, the path migration control unit 111 provided in the new CU 11 accesses the old CU 13 via the third access paths 30 and 31 of the access from the second access paths 20 'and 21' side. By the relay of CP
The first access route 20 and the first access route 21 from U10 are the second access route 20 'and the second access route 2
After all the connections were switched to the new CU11 as 1 ',
If the data migration control unit 112 is instructed to start the data migration, and further the path migration control unit 111 is designated to be processed by its own CU in synchronization with this, all the data can be migrated without interruption, and the non-stop data Transition can be implemented.

The entire processing procedure of subsystem migration including data migration in this embodiment is shown in FIG.
The flow chart illustrated in FIG.

That is, in steps 101 to 106, the first access routes 20 and 21 are changed to the second access route 20 ',
After the path switching operation for switching to 21 'is executed, the data is copied (migrated) from the old CU 13 side to the new CU 11 side via the third access paths 30 and 31 (step 1).
07) is executed for all the data required to be migrated in the old VOL 14 under the old CU 13 (step 10).
8), then the third access routes 30, 31 and old C
Remove U13 and old VOL14 (Step 10)
9).

Here, the processing of the access request from the CPU 10 during the data migration in the above steps 107 to 108 is as shown in the flowcharts illustrated in FIGS. 8 and 9 as an example.

That is, the copy process as illustrated in FIG. 9 is executed as the background process, and the access request process illustrated in FIG. 8 is executed at any time during this process.

First, in the copy process of FIG. 9, when data is copied in track units as an example, a bitmap for copy management (not shown) is referred to (step 30).
1) The presence or absence of uncopied tracks in the old VOL 14 is checked (step 302), and if there is, the one with the smallest track number in the uncopied tracks is selected (step 30).
3), the new VOL via the third access routes 30 and 31
After copying to the 12 side (step 304), the operation of updating the bitmap for copy management to the track copy completed (step 305) is performed for all tracks to be transferred.

On the other hand, as illustrated in FIG. 8, the new CU1
In No. 1, when a command is received from the CPU 10 (step 201), it is checked whether or not the access area of the command is an uncopied area (step 202). If the command is an uncopied area, the command is a read-related command. It is checked (step 203). In the case of a read command, the old VOL1 is passed via the third access routes 30 and 31.
After copying the track including the data to be accessed from the side of 4 to the new VOL 12 (step 205), the bitmap for copy management is updated to copied (step 20).
6) After that, command processing is executed (step 20).
7).

On the other hand, when it is determined in step 202 that the copy area is accessed, the command processing is immediately executed in step 207.

If it is determined in step 203 that the command is not a read command (that is, a write command), it is determined whether the old data is a required write command (step 20).
4), if necessary, execute step 205 and subsequent steps,
If unnecessary, the command processing of step 207 is executed.

By such processing, it is possible to shift from the old subsystem to the new subsystem without stopping the information processing system, and the second access of the first access routes 20 and 21 which is in non-stop operation. Route 2
The data migration processing after switching to 0'and 21 'can be performed smoothly and appropriately.

(Embodiment 2) FIG. 2 is a conceptual diagram showing another embodiment of the information processing system in which the subsystem migration method of the present invention is implemented, and FIG. 10 and FIG.
Is a flowchart showing an example of the operation.

As an example of the configuration of the information processing system of the present embodiment, the new CU 11 does not include the path migration control unit and the data migration control unit, and the old CU 13 does not include the path migration control unit 13.
1 and a point that the data migration control unit 132 is provided.
Different from.

First, the schematic operation of the route switching operation by relaying the access request generated from the first access routes 20 and 21 to the old CU 13 to the new CU 11 side via the third access routes 30 and 31 will be described.

Since the normal process is not related to the path transfer, the processing by the own CU is designated in the path transfer control unit 131. First, the third access routes 30 and 31 between the new CU 11 and the old CU 13 are newly established (step 401, step 402). Further, another CU is set in the path transfer control unit 131.
Specify the process in. Thus, the old CU 13 does not process it, but the new CU 11 processes it only by relaying it through the third access routes 30 and 31.

Next, the connection of the first access route 20 to the second access route 20 'is switched. At this time, the CPU 10
Therefore, the first access route 20 is set to the offline state, and the access of the CPU 10 using the route is stopped. Further, the third access route 30 is removed (steps 405 and 406). When connection is over, CPU1
From 0, the second access route 20 ′ (original first access route 20) is brought into the online state.

As the access to the old CU 13 from the CPU 10, the first access route 2 that has been used until then
The access to the old CU 13 side using 1 and the access to the new CU 11 using the second access route 20 ′ are started. New C for access using the second access route 20 '
Although the U11 receives, the access that the new CU11 received through the third access route 30 is received by the second access route 2
The processing is continued as it is (step 407). The access to the old CU13 side using the first access route 21 is the third access route 3
It is relayed to the new CU 11 via 0 and 31 and processed (step 408).

In the same procedure, the first access route 21 is set to the second access route 21.
Switch to access route 21 ', and change to third access route 31
To remove. In this way CP without stopping access
The access route from U10 to the old CU13 is changed to the new CU11.
Connection can be switched to (Step 40
9). Then, the old CU 13 and the old VOL 14 under its control are removed (step 410).

However, here, when applied to the migration of a storage subsystem that requires data migration, in the CU of the storage subsystem that controls access to VOL data, if the data has not been migrated to the new CU 11. , C
The access from the PU 10 cannot be processed. Therefore, in the present embodiment, first, when the third access routes 30 and 31 are newly established, the data migration control unit 132 is instructed to start data migration (step 403). At the time when all the data can be transferred and the new CU 11 can be processed even if the new CU 11 is directly accessed from the CPU 10 (step 40
4) By executing the path switching process after step 405 and switching the connection without stopping the access from the CPU 10, non-stop data migration can be implemented.

Here, an example of an access request (command processing) generated from the CPU 10 during the path switching operation of steps 405 to 409 after the completion of the data migration of steps 403 to 404 will be described with reference to FIG.

That is, when the old CU 13 receives a command from the CPU 10 (step 501), the old CU 13 checks whether the access area of the command is a copied area (step 502). Check whether is a read command (step 5
03), if the command is not a read command (write command), command processing is executed by both the new CU 11 and the old CU 13 (step 504). On the other hand, step 503
If the command is determined to be a read command, the old CU13
Performs command processing using the data of the old VOL 14 (step 505).

If it is determined in step 502 that the access is to an uncopied area, in step 505
The old CU 13 uses the data of the old VOL 14 to perform command processing.

By such processing, the information processing system is stopped by relaying the access request to the new CU 11 side via the third access routes 30 and 31 by the path migration control unit 131 provided on the old CU 13 side. Without
The old subsystem can be migrated to the new subsystem, and the path switching operation during data migration can be accurately executed in a non-stop state.

(Third Embodiment) FIG. 3 is a conceptual diagram showing another embodiment of the information processing system in which the subsystem migration method of the present invention is implemented. The configuration of the information processing system in the present embodiment differs from that in FIG. 1 only in that the old CU 13 includes a path migration control unit 131. Here, the characteristic outline operation in the present embodiment will be explained, and in the flow of the explanation, the third access route 30 by the path migration control unit 111 and the path migration control unit 131 to the other subsystems from each other. An outline operation of a path switching operation by relaying an access request via the access terminals 31 and 31 will be described.

Since the normal process is not related to the path migration, the path migration control unit 111 and the path migration control unit 131 have their own CU.
Processing is specified. First, the new CU11 and the old CU
Third access routes 30 and 31 between 13 are newly established.
Here, the path transfer control unit 111 and the path transfer control unit 131
When the access is received from the CPU 10, it is processed by the own CU, and the other CU is processed through the third access routes 30 and 31.
Specified to perform the same access to
This is called processing in both CUs). Old CU13 to new CU
The access to 11 is an error because there is no data in the new CU 11 yet, but there is no problem because the old CU 13 is performing the processing from the CPU 10.

Next, the data migration control unit 112 is instructed to start data migration. At this point, the new CU 11 is the old CU
The access from 13 is also normally processed in the case of the data migrated portion, and is normally processed in the case of the data non-migrated portion because the data is read from the migration source by the conventional data migration function. After this, during the data transfer, the first
The connection of the access route 20 is switched to the second access route 20 '. At this time, the first access route 20 from the CPU 10 is set to the offline state, and the C
The access of PU10 is stopped. C after connection
The second access route 20 ′ (original first access route 20) is brought online from the PU 10. As the access to the old CU 13, the CPU 10 starts the access using the second access route 20 ′ together with the access using the first access route 21 that has been performed so far.
New CU11 for access using the second access route 20 '
Received, but the old C that had been received until then in the new CU11
It is processed together with the access from U13. However,
Since the access is from the CPU 10, the same access is made to the old CU 13. In this state, the CPU 10 to the old C
Both the U13 and the new CU11 receive access and perform processing, but since all access is also made to the partner CU, the input / output processing is interrupted and the second access route 20 'and the first access route 21 are switched and accessed. Can also be processed.

In the same procedure, the first access route 21 is set to the second access route 21.
Switch to the access route 21 '. In this way, the old CU
13 and the new CU 11 are both path migration control units 131
And the path migration control unit 111, and the old CU 13 from the CPU 10 is maintained without stopping the access of the CPU 10.
It is possible to switch the connection of the first access paths 20, 21 to the new CU 11 to the second access paths 20 ', 21' to the new CU 11.

After that, data migration (copy) of all data to be migrated from the old CU 13 side to the new CU 11 side is performed by the conventional data migration function by the data migration control unit 112 provided on the new CU 11 side. To be done. In this way, even during data migration, data migration can be performed without stopping access of the CPU 10 and connection switching of the first access paths 20 and 21 to the second access paths 20 ′ and 21 ′.

(Embodiment 4) FIG. 4 is a conceptual diagram showing an example of the configuration and operation of another embodiment of the information processing system in which the subsystem migration method of the present invention is implemented. The configuration of the information processing system in this embodiment is
A path switching device 1 is provided with a plurality of CPUs 10a and 10b instead of one CPU 10, and the CPU 10a dynamically switches the access paths by the access paths 20a and 21a and the CPU 10b by the access paths 20b and 21b.
5, the path switching device 15 and the old CU1
1 is different in that points 3 are connected by a first access path 20c and a first access path 21c.

Further, in the subsystem migration processing according to this embodiment, the first access path 20c and the first access path 21c are the second access path 20c between the path switching device 15 and the new CU 11 which is the data migration destination. The connection is changed as'and the second access route 21c '.

In the present embodiment, the access from the CPU 10a through the access route 20a is performed by the old CU through the path switching device 15 and the first access route 20c.
It is implemented in 13. Similarly, the first access route 21c from the access route 21a, the first access route 20c from the access route 20b in the case of the CPU 10b,
From the access route 21b to the old CU 13 through the first access route 21c. In such a case, the first access route 20c is physically one route, but has two logical access routes so that the access from the access route 20a and the access from 20b can be distinguished.

Similarly, in the old CU 13 as well, each of the second access routes 20c '(second access route 21c') corresponds to the access route 2 on the CPU 10a (CPU 10b) side.
0a (access route 20b) and access route 21a
It functions as two logical access routes corresponding to each (access route 21b), and these two routes are recognized and processed as access routes from different CPUs 10a and 10b.

As illustrated in FIG. 1, the path transition control unit 111 is provided on the new CU 11 side to relay the access request from the second access route to the old CU 13 side via the third access routes 30 and 31. Therefore, the access route switching operation from the first to the second access route is performed in the non-stop state, and the data migration is performed by the same procedure as that performed after the completion of the route switching, and the first access route 20c is switched to When the route is set to 20c ′, it is recognized in the new CU 11 as well.

That is, the access received from the two logical access paths is divided, and the CPU 10a or 10b determines which of the CPUs the access is to be executed. The process during the switching of the access route means that the path migration control unit 111 receives the access received from the CPU.
Although the same access is made by relaying to the CPU, it is naturally necessary to divide the old CU 13 from which CPU the access is made from. Since this division is made according to the difference of access routes, the third access route 3 is used in this embodiment.
This is realized by selectively using 1 for access from the CPU 10a and 3rd access route 30 for access from the CPU 10b. Similarly, when the first access route 21c is switched to the second access route 21c ', the same process may be performed. In this way, the CP connected to the old CU13
When there are multiple U, at least as many new CUs as there are CPUs
It is possible to perform the inter-subsystem path migration by providing the third access route between the 1 and the old CU 13 and realize the data migration by the present embodiment illustrated in FIG.

In the present embodiment, the third access route 30 is used for accessing from the CPU 10a and the third access route 31 is used for accessing from the CPU 10b. However, only the third access route 30 can be used. When there is no such case, that is, when the number of access routes between the new CU 11 and the old CU 13 is smaller than the number of CPUs, it is also possible to provide two logical access routes in the third access route 30 and use them properly. Needless to say, a plurality of logical access paths can be provided and used even when the number of third access paths is sufficient (when the number of CPUs is greater than or equal to the number of CPUs). Furthermore, the second access route 20c 'is associated with the third access route 30, the second access route 21c' is associated with the third access route 31, and the second access route 20c 'and the second access route 21c' are associated with each other. It is also possible to provide a certain logical access route as it is on the third access routes 30 and 31.

(Embodiment 5) Another embodiment of the subsystem migration method of the present invention will be described with reference to FIGS. FIG. 5 is a conceptual diagram showing an example of an old CU device information table 40 in which device information of the old CU 13 is stored in a storage unit such as a buffer memory provided in the new CU 11 of the information processing system illustrated in FIG. is there.

Normally, in a subsystem that operates under the control of a CPU in an information processing system, information such as the device configuration of the subsystem is sent to the CPU as necessary in order to determine the environment and specifications of the subsystem. Command interface for reading to the side (device information input request)
Is equipped with.

In response to the device information input request from the CPU 10, while the old CU 13 is connected, the same request is issued to the old CU 13 and if the returned information is input to the CPU 10, the CPU 10 may find an access route failure or the like. It is possible to continue using it as a subsystem without judgment. However, the old CU 13 is usually removed after the data migration is completed. Therefore, in the present embodiment, the new CU 11 is previously provided with the third access routes 30, 31 so that the device information of the old CU 13 can be responded (input) to the CPU 10 even after the removal.
All device information input requests are sent to the old CU 13 via the, and the returned information is entered in the old CU device information table 40 with the input request name 40a and the information 40b returned from the old CU 13 in response to the input request. And record it, and the subsequent CPU 10
Device information input request from the old CU device information table 40
The information 40b is read and responded.

In this way, even when the CPU 10 is a CPU that compares past device information, after removing the old CU 13,
It is possible to continue using the new CU11. That is, the old CU 13 can be removed without causing a concern such as system down. Further, in this embodiment, the new CU 11
Although the storage means such as the old CU device information table 40 provided in the above is used, the device information originally possessed by the new CU 11 may be rewritten to the information of the old CU 13.

As described above, in the present invention, the CPU
Since you can switch the path from the old subsystem of the migration source to the new subsystem of the migration source without stopping access at all,
A completely non-disruptive system migration is possible.

Further, in the disk subsystem,
Even if the new disk subsystem at the migration destination is equipped only with the function of performing path switching without stopping CPU access, complete non-disruptive data migration including path switching can be performed.

In the disk subsystem, C
Even if the old disk subsystem of the migration source has only the function to switch paths without stopping PU access,
Complete non-stop data migration including path switching is possible.

In the disk subsystem, C
A function that can switch paths without stopping PU access,
By providing both the new subsystem of the migration destination and the old disk subsystem of the migration source, it is possible to perform completely non-disruptive data migration that enables path switching during data migration.

Furthermore, even in a subsystem operating under the control of a plurality of CPUs, a completely non-stop system transfer can be performed.

Further, even in the case of a CPU which stores the device information of the subsystem and compares it with the newly read device information of the current subsystem, the device information of the old subsystem is also updated. The subsystem side reads and saves the device information in advance and responds to the device information of the old subsystem that was saved in response to the device information input request. The system can be quickly removed.

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 embodiments, and various modifications can be made without departing from the scope of the invention. Needless to say.

For example, the storage subsystem accompanied by data migration has been described as an example of the subsystem.
Not limited to this, it can be widely applied to general subsystems that do not require data migration.

[0085]

According to the subsystem migration method of the present invention, it is possible to continue the access from the host device to the subsystem side even during the switching from the old subsystem to the new subsystem. .

Further, there is no need to stop the access from the upper device to the subsystem side due to the data transfer procedure, and the data transfer can be performed in the non-stop state.

Further, there is an effect that the old subsystem operating under the control of a plurality of host devices can be smoothly transferred to the new subsystem under non-stop operation.

Further, it is possible to avoid the occurrence of a failure due to the environmental change such as the device information accompanying the transition from the old subsystem to the new subsystem, and to perform the smooth subsystem migration.
The effect is obtained.

[Brief description of drawings]

FIG. 1 is a conceptual diagram showing an example of the configuration and operation of a general-purpose computer system which is an embodiment of an information processing system in which a subsystem migration method of the present invention is implemented.

FIG. 2 is a conceptual diagram showing another embodiment of the information processing system in which the subsystem migration method of the present invention is implemented.

FIG. 3 is a conceptual diagram showing another embodiment of the information processing system in which the subsystem migration method of the present invention is implemented.

FIG. 4 is a conceptual diagram showing an example of the configuration and operation of another embodiment of the information processing system in which the subsystem migration method of the present invention is implemented.

FIG. 5 is a conceptual diagram showing an example of contents of a storage unit used in another embodiment of the subsystem migration method of the present invention.

FIG. 6 is a flowchart showing an example of the operation of the information processing system in which the subsystem migration method of the present invention is implemented.

FIG. 7 is a flowchart showing an example of the operation of the information processing system in which the subsystem migration method of the present invention is implemented.

FIG. 8 is a flowchart showing an example of the operation of the information processing system in which the subsystem migration method of the present invention is implemented.

FIG. 9 is a flowchart showing an example of the operation of the information processing system in which the subsystem migration method of the present invention is implemented.

FIG. 10 is a flowchart showing an example of the operation of another embodiment of the information processing system in which the subsystem migration method of the present invention is implemented.

FIG. 11 is a flowchart showing an example of the operation of another embodiment of the information processing system in which the subsystem migration method of the present invention is implemented.

[Explanation of symbols]

5a to 5i ... Operation, 10, 10a, 10b ... CPU, 1
1 ... New disk controller unit (new CU), 12
… New disk volume (new VOL), 12a… Device path, 13… Old disk controller unit (old C
U), 14 ... Old disk volume (old VOL), 14
a ... Device path, 15 ... Path switching device, 20a,
20b ... Access route, 21a, 21b ... Access route, 20 ... First access route, 20 '... Second access route, 20c ... First access route, 20c' ... Second access route, 21 ... First access route, 21 '... second access route, 21c ... first access route, 21c' ... second access route, 30, 31 ... third access route, 40 ... old CU device information table, 40a ... input request name, 40b ... information,
50 ... CPU access path, 51 ... Path transfer control unit designation, 52 ... Process CU, 111 ... Path transfer control unit, 112
... Data migration control unit, 131 ... Path migration control unit, 132
... Data migration control unit.

Continued front page    (72) Inventor Katsunori Nakamura             2880 Kozu, Odawara City, Kanagawa Stock Association             Storage Systems Division, Hitachi, Ltd. F-term (reference) 5B065 BA01 CA11 CE22 EA23

Claims (5)

[Claims] 1. A first subsystem having a first access path for receiving an access request for data issued by a higher-level device, and a second access for receiving an access request for data issued by the higher-level device. A storage system comprising a second subsystem having a path, wherein the storage system includes a first access path and a third access path connecting the second subsystem, The first subsystem is the first device from the first device.
The storage system for relaying an access request received via the access path to the second subsystem via the third access path. 2. When the first subsystem receives an access request from the higher-level device, the first subsystem does not perform the processing relating to the access request,
The storage system according to claim 1, wherein the access command is relayed to the second subsystem via the third access path. 3. A first subsystem having a first access path for receiving an access request for data issued by a higher-level device, and a second access for receiving an access request for data issued by the higher-level device. A storage system comprising a second subsystem having a path, wherein the storage system includes a first access path and a third access path connecting the second subsystem, The first subsystem has a first migration control unit that specifies whether to process the access request by the first control device or the second subsystem, and the first subsystem The system uses the first device to the first device.
When the access request is received via the access path of the first sub-system, the first sub-controller controls the second sub-system based on the designation of the first migration control unit. A storage system characterized by transmitting the access request to the second subsystem via the third access path without processing the access request in the subsystem. 4. The transmitted access request is the first access request.
4. The storage system according to claim 3, wherein the storage system relates to input / output processing equivalent to an access request received by the subsystem. 5. The first migration control means, when relaying a command to the second subsystem via the third access path, performs an operation equivalent to that of the host device. The storage system according to claim 3, wherein the storage system is a storage system.