INFORMATION PROCESSING SYSTEM AND METHOD OF CONTROLLING THE SAME

• US 20130117223A1
• Filed: 11/09/2011
• Published: 05/09/2013
• Est. Priority Date: 11/09/2011
• Status: Active Grant

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Niki, Yosuke
• Granted Patent: US 8,688,632 B2
• Time in Patent Office: Days
• US Class Current: 707/610
• CPC Class Codes: G06F 16/185 Hierarchical storage management [HSM] systems, e.g. file migration or policies thereof details of archiving G06F16/11

Abstract

In an information processing system 1 including first to third storage systems 50a to 50c, the first storage system 50a performs stubbing by managing entity data of a file in the second storage apparatus 10b, migration by replicating the data of the file to the second storage apparatus 10b, and recall by writing the entity data of the file from the second storage apparatus 10b back to the first storage apparatus 10a. Moreover, the second storage system 50b performs remote copy by replicating metadata of the migrated file from the first storage apparatus 10a to the third storage apparatus 10c at timing of the migration (synchronously), and replicating the entity data of the file to the third storage apparatus 10c at timing independent of the timing of the migration (asynchronously).

Claims

1. An information processing system comprising:

a first storage system that performs processing for a file stored in a first storage apparatus in response to an I/O request received from a client apparatus;

a second storage system communicatively coupled to the first storage system and performs processing for a file stored in a second storage apparatus; and

a third storage system communicatively coupled to the second storage system and performs processing for a file stored in a third storage apparatus, wherein the first storage system performs

stubbing by storing metadata of data in a file, stored in the first storage apparatus, into the first storage apparatus and storing entity data of the data in the file into the second storage apparatus, and deleting the entity data from the first storage apparatus to manage the file,

migration by replicating the data of the file stored in the first storage apparatus into the second storage apparatus to perform the stubbing, and

recall by writing the entity data from the second storage apparatus back to the first storage apparatus when receiving the I/O request requiring the entity data from the client apparatus, and

the second storage system performs

remote copy by replicating the metadata in the file replicated from the first storage apparatus by the migration to the third storage apparatus at timing of the migration, and replicating the entity data in the file replicated from the first storage apparatus by the migration to the third storage apparatus at a timing independent of the timing of the migration.

2. The information processing system according to claim 1, wherein

the first storage system performs the stubbing of the file corresponding to the entity data, when replication of the entity data to the third storage apparatus by the remote copy is completed.

3. The information processing system according to claim 2, wherein

the first storage system notifies the second storage system of information specifying a file to be subjected to the stubbing,

the second storage system replicates the entity data in the file from the second storage apparatus to the third storage apparatus upon receipt of the notification, and notifies the first storage system of completion of the replication, and

the first storage system performs the stubbing of the file corresponding to the entity data upon receipt of the notification.

4. The information processing system according to claim 2, wherein

the third storage system manages difference information that is information indicating whether or not metadata and entity data in a file stored in the third storage apparatus are consistent, and

the first storage system acquires the difference information from the third storage system, and determines based on the acquired difference information whether or not the replication of the entity data to the third storage apparatus by the remote copy is completed.

5. The information processing system according to claim 1, wherein

the third storage system manages difference information that is information indicating whether or not metadata and entity data in a file stored in the third storage apparatus are consistent, and

when a failure occurs in the second storage system,

the third storage system requests the first storage system for entity data of a file of which the replication of the entity data to the third storage apparatus in the remote copy is not completed, specified based on the difference information,

the first storage system transmits the entity data to the third storage system in response to the request, and

the first storage system performs failover from the second storage system to the third storage system when the third storage system has no more file having entity data yet to be replicated to the third storage apparatus by the remote copy.

6. The information processing system according to claim 1, wherein the first storage system

sets a stubbing priority for each file stored in the first storage apparatus, the stubbing priority being information indicating a priority order considered in stubbing,

performs the stubbing of the files according to the stubbing priority, and

sets the stubbing priority so that the files are stubbed in descending order of file size.

7. The information processing system according to claim 1, wherein the first storage system

sets a stubbing priority for each file stored in the first storage apparatus, the stubbing priority being information indicating a priority order considered in stubbing,

performs the stubbing of the files according to the stubbing priority, and

sets the stubbing priority so that the files are stubbed in ascending order of access frequency from the client apparatus.

8. The information processing system according to claim 1, wherein the first storage system

sets a stubbing priority for each file stored in the first storage apparatus, the stubbing priority being information indicating a priority order considered in stubbing,

performs the stubbing of the files according to the stubbing priority, and

sets the stubbing priority so that the files are stubbed in ascending order of file significance.

9. The information processing system according to claim 1, wherein the first storage system

sets a stubbing priority for each file stored in the first storage apparatus, the stubbing priority being information indicating a priority order considered in stubbing,

performs the stubbing of the files according to the stubbing priority, and

sets the stubbing priority so that the files are stubbed in descending order of file significance.

10. The information processing system according to claim 6, wherein

the first storage system performs stubbing of a plurality of files in chronological order of final update date/time of the files when there are a plurality of files having the same stubbing priority in the files stored in the first storage apparatus.

11. The information processing system according to claim 6, wherein the first storage system

increases the stubbing priority of a file having a file size exceeding a predetermined threshold, and

resets the threshold to be lower within a predetermined range when a remaining capacity exceeding a predetermined remaining capacity cannot be reserved in the first storage apparatus even though the stubbing is executed.

12. The information processing system according to claim 1, wherein

a plurality of the second storage systems, to be data replication destinations of the file stored in the first storage apparatus in the migration, exists for the first storage system as a single unit, and

the first storage system

sets a stubbing priority for each file stored in the first storage apparatus, the stubbing priority being information indicating the priority order considered in stubbing,

performs the stubbing of the files according to the stubbing priority,

manages a weight coefficient set for each of the second storage systems, and

sets the stubbing priority of the files according to the weight coefficients of the second storage systems that perform processing in response to the I/O request for a file stored in the second storage apparatus serving as data replication destination of the file stored in the first storage apparatus in the migration.

13. The information processing system according to claim 1, wherein

the information processing system includes one or more other first storage systems different from the first storage system, that provides a client apparatus with a file stored in the first storage apparatus,

the first storage system sets a stubbing priority for each file stored in the first storage apparatus, the stubbing priority being information indicating a priority order considered in stubbing, and performs the stubbing of the files according to the stubbing priority, and

the other first storage system restores the entity data of the file in the first storage apparatus, replicated to the second storage apparatus by the migration, to another first storage apparatus that is the first storage apparatus of its own by performing the recall in a processing of the I/O request received from the client apparatus by the other first storage system,

the second storage system manages redundancy of data of a file in the information processing system,

the first storage system notifies the second storage system of information specifying the stubbed file in the stubbing of the file stored in the first storage apparatus,

the second storage system reduces the redundancy of the data of the stubbed file specified by the information when receiving the notification,

the second storage system increases the redundancy of the file when the recall is performed for the file stored in the first storage apparatus, and

the first storage system acquires the redundancy of the file stored in the first storage apparatus from the second storage system, and sets the stubbing priority of the file stored in the first storage apparatus based on the acquired redundancy.

14. The information processing system according to claim 13, wherein

the first storage system and the other first storage system are communicatively coupled to each other,

the other first storage system sets another stubbing priority for each file stored in the first storage apparatus, the other stubbing priority being information indicating a priority order considered in stubbing, and performs the stubbing of the files according to the stubbing priority,

the second storage system notifies the other first storage system that the other stubbing priority needs to be updated when the redundancy is updated, and

the other first storage system acquires the stubbing priority managed by the first storage system from the first storage system upon receipt of the notification, and updates the other stubbing priority managed by itself.

15. The information processing system according to claim 1, wherein

the first storage system performs the stubbing of a file corresponding to the entity data when replication of the entity data to the third storage apparatus by the remote copy is completed,

the first storage system notifies the second storage system of information specifying a file to be subjected to the stubbing,

the second storage system replicates entity data of the file from the second storage apparatus to the third storage apparatus upon receipt of the notification, and notifies the first storage system of completion of the replication,

the first storage system performs the stubbing of a file corresponding to the entity data upon receipt of the notification,

the third storage system manages difference information that is information indicating whether or not metadata and entity data in a file stored in the third storage apparatus are consistent, and

the first storage system acquires the difference information from the third storage system, and determines based on the acquired difference information whether or not the replication of the entity data to the third storage apparatus by the remote copy is completed,

when a failure occurs in the second storage system,

the third storage system requests the first storage system for the entity data of a file of which the replication of the entity data to the third storage apparatus in the remote copy is not completed, specified based on the difference information,

the first storage system transmits the entity data to the third storage system in response to the request, and

performs failover from the second storage system to the third storage system when the third storage system has no more file having entity data yet to be replicated to the third storage apparatus by the remote copy,

the first storage system

sets a stubbing priority for each file stored in the first storage apparatus, the stubbing priority being information indicating a priority order considered in stubbing,

performs the stubbing of the files according to the stubbing priority,

sets the stubbing priority so that the files are stubbed in descending order of file size, sets the stubbing priority so that the files are stubbed in ascending order of access frequency from the client apparatus, sets the stubbing priority so that the files are stubbed in ascending order of file significance, or sets the stubbing priority so that the files are stubbed in descending order of file significance,

performs stubbing of a plurality of files in chronological order of final update date/time of the files when there are a plurality of files having the same stubbing priority in the files stored in the first storage apparatus,

increases the stubbing priority to a file having a file size exceeding a predetermined threshold when the stubbing priority is set so that the files are stubbed in descending order of file size, resets the threshold to be lower within a predetermined range when a remaining capacity exceeding a predetermined remaining capacity cannot be reserved in the first storage apparatus even though the stubbing is executed, when the stubbing priority is set so that the stubbing is performed in descending order of file size,

when, a plurality of the second storage systems, to be data replication destinations of the file stored in the first storage apparatus in the migration, exists for the first storage system as a single unit,

the first storage system

manages a weight coefficient set for each of the second storage systems, and

sets the stubbing priority of the files according to the weight coefficients of the second storage systems that perform processing in response to the I/O request for a file stored in the second storage apparatus serving as data replication destination of the file stored in the first storage apparatus in the migration,

the information processing system includes one or more other first storage systems different from the first storage system, that provides a client apparatus with a file stored in the first storage apparatus,

the first storage system sets a stubbing priority for each file stored in the first storage apparatus, the stubbing priority being information indicating a priority order considered in stubbing, and performs the stubbing of the files according to the stubbing priority, and

the other first storage system restores the entity data of the file in the first storage apparatus, replicated to the second storage apparatus by the migration, to another first storage apparatus that is the first storage apparatus of its own by performing the recall in a processing of the I/O request received from the client apparatus by the other first storage system,

the second storage system manages redundancy of data of a file in the information processing system,

the first storage system notifies the second storage system of information specifying the stubbed file in the stubbing of the file stored in the first storage apparatus,

the second storage system reduces the redundancy of the data of the stubbed file specified by the information when receiving the notification,

the second storage system increases the redundancy of the file when the recall is performed for the file stored in the first storage apparatus, and

the first storage system acquires the redundancy of the file stored in the first storage apparatus from the second storage system, and sets the stubbing priority of the file stored in the first storage apparatus based on the acquired redundancy,

the first storage system and the other first storage system are communicatively coupled to each other,

the other first storage system sets another stubbing priority for each file stored in the first storage apparatus, the stubbing priority being information indicating a priority order considered in stubbing, and performs the stubbing of the files according to the stubbing priority,

the second storage system notifies the other first storage system that the other stubbing priority needs to be updated when the redundancy is updated, and

the other first storage system acquires the stubbing priority managed by the first storage system from the first storage system upon receipt of the notification, and updates the other stubbing priority managed by itself.

16. A method of controlling an information processing system including

a first storage system that performs processing for a file stored in a first storage apparatus in response to an I/O request received from a client apparatus,

a second storage system communicatively coupled to the first storage system and performs processing for a file stored in a second storage apparatus, and

a third storage system communicatively coupled to the second storage system and performs processing for a file stored in a third storage apparatus,

the method comprising the steps of:

the first storage system performing

stubbing by storing metadata of data in a file, stored in the first storage apparatus, into the first storage apparatus and storing entity data of the data in the file into the second storage apparatus, and deleting the entity data from the first storage apparatus to manage the file;

migration by replicating the data of the file stored in the first storage apparatus into the second storage apparatus to perform the stubbing; and

recall by writing the entity data from the second storage apparatus back to the first storage apparatus when receiving the I/O request requiring the entity data from the client apparatus, and

the second storage system performs

remote copy by replicating the metadata in the file replicated from the first storage apparatus by the migration to the third storage apparatus at timing of the migration, and replicating the entity data in the file replicated from the first storage apparatus by the migration to the third storage apparatus at timing independent of a timing of the migration.

17. The method for controlling an information processing system, according to claim 16, wherein

the first storage system performs the stubbing of the file corresponding to the entity data, when replication of the entity data to the third storage apparatus by the remote copy is completed.

18. The method for controlling an information processing system, according to claim 17, wherein

the first storage system notifies the second storage system of information specifying a file to be subjected to the stubbing,

the second storage system replicates the entity data in the file from the second storage apparatus to the third storage apparatus upon receipt of the notification, and notifies the first storage system of completion of the replication, and

the first storage system performs the stubbing of the file corresponding to the entity data upon receipt of the notification.

19. The method for controlling an information processing system, according to claim 17, wherein

the third storage system manages difference information that is information indicating whether or not the metadata and entity data in a file stored in the third storage apparatus are consistent, and

the first storage system acquires the difference information from the third storage system, and determines based on the acquired difference information whether or not the replication of the entity data to the third storage apparatus by the remote copy is completed.

20. The method for controlling an information processing system, according to claim 16, wherein

the third storage system manages difference information that is information indicating whether or not metadata and entity data, in a file stored in the third storage apparatus, are consistent, and

when a failure occurs in the second storage system,

the third storage system requests the first storage system for entity data of a file of which the replication of the entity data to the third storage apparatus in the remote copy is not completed, specified based on the difference information,

the first storage system transmits the entity data to the third storage system in response to the request, and

the first storage system performs failover from the second storage system to the third storage system when the third storage system has no more file having entity data yet to be replicated to the third storage apparatus by the remote copy.