Storage subsystem and storage subsystem control method

• US 7,739,454 B2
• Filed: 06/12/2007
• Issued: 06/15/2010
• Est. Priority Date: 02/05/2004
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Hoshino, Sachiko, Sakaguchi, Takashi, Sugino, Shoji, Nagasoe, Yasuyuki
• Primary Examiner(s): BATAILLE, PIERRE MICHE
• Application Number:
  US11/808,606
• Publication Number:
  US 20070245085A1
• Time in Patent Office: 1,099 Days
• Field of Search: 711/113, 711/114, 711/170, 711/129, 711/173
• US Class Current: 711/129
• CPC Class Codes: G06F 12/084 with a shared cache
• CPC Class Codes: G06F 12/0866 for peripheral storage systems, e.g. disk cache
• CPC Class Codes: G06F 12/0871 Allocation or management of cache space
• CPC Class Codes: G06F 2003/0697 device management, e.g. handlers, drivers, I/O schedulers
• CPC Class Codes: G06F 2212/282 Partitioned cache
• CPC Class Codes: G06F 2212/311 In host system
• CPC Class Codes: G06F 2212/312 In storage controller
• CPC Class Codes: G06F 2212/6042 Allocation of cache space to multiple users or processors
• CPC Class Codes: G06F 3/0601 Interfaces specially adapted for storage systems
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

The present invention partitions a cache region of a storage subsystem for each user and prevents interference between user-dedicated regions. A plurality of CLPR can be established within the storage subsystem. A CLPR is a user-dedicated region that can be used by partitioning the cache region of a cache memory. Management information required to manage the data stored in the cache memory is allocated to each CLPR in accordance with the attribute of the segment or slot. The clean queue and clean counter, which manage the segments in a clean state, are provided in each CLPR. The dirty queue and dirty counter are used jointly by all the CLPR. The free queue, classification queue, and BIND queue are applied jointly to all the CLPR, only the counters being provided in each CLPR.

Claims

1. A storage subsystem, comprising:

a plurality of channel adapters that each controls the exchange of data with a host device;

a plurality of storage device groups that each provides a logical storage region;

a plurality of disk adapters that each controls the exchange of data with each of the storage device groups;

a cache memory that is used by each of the channel adapters and each of the disk adapters;

a plurality of cache partition regions constituted by logically partitioning the cache memory; and

a control memory that stores management information for managing each of the cache partition regions,

wherein the management information is constituted by both a plurality of partition management information that is used to independently manage corresponding ones of the cache partition regions, and common management information that is used to commonly manage a plurality of the cache partition regions,

wherein when a cache partition region corresponding to a certain logical storage region is transferred from a first cache partition region to a second cache partition region, information in relation to a queue and counter included in the management information stored in the control memory is updated,

wherein the management information is constituted by sub-management information of a plurality of types, and

wherein the partition management information are constituted by partitioning, based on an indication of a kind of status in which data is stored in an area of the cache partition region, some of the sub-management information items for each of the cache partition regions, the remainder of the sub-management information being used as the common management information.

2. The storage subsystem according to claim 1, wherein the partition management information and the common management information are established based on an indication of a kind of status in which data is stored in an area of the cache partition region.

3. The storage subsystem according to claim 1, wherein each of the cache partition regions can be established for each of the channel adapters.

4. The storage subsystem according to claim 1, wherein one cache partition region among the cache partition regions is established as a common region and a new cache partition region is established by allocating resources belonging to the common region.

5. A storage system according to claim 1, wherein the information in relation to the queue and counter which is to be updated is both queue and counter information in relation to the first cache partition region and queue and counter information in relation to the second cache partition region.

6. A storage system according to claim 1, wherein the information in relation to the queue which is to be updated includes connecting a movement target slot control block table to a movement-destination queue of the second cache partition region.

7. A storage subsystem, comprising:

a plurality of channel adapters that each controls the exchange of data with a host device;

a plurality of storage device groups that each provides a logical storage region;

a plurality of disk adapters that each controls the exchange of data with each of the storage device groups;

a cache memory that is used by each of the channel adapters and each of the disk adapters;

a plurality of cache partition regions constituted by logically partitioning the cache memory; and

a control memory that stores management information for managing each of the cache partition regions,

wherein the management information is constituted by partition management information items corresponding to the cache partition regions, and common management information that is applied to all of the cache partition regions,

wherein the management information is constituted by queues of a plurality of types, and counters that are associated with each of the queues,

wherein the partition management information items are constituted by providing some of the queues and counters in each of the cache partition regions based on an indication of a kind of status in which data is stored in an area of the cache partition region, the remainder of the queues and the remainder of the counters being used as the common management information,

wherein when a cache partition region corresponding to a certain logical storage region is transferred from a first cache partition region to a second cache partition region, information in relation to a queue and counter included in the management information stored in the control memory is updated, and

wherein the information in relation to the counter which is to be updated includes subtracting a count from an in-use counter and a counter in the first cache partition region in accordance with a number of slots moved.

8. The storage subsystem according to claim 7, wherein one of a first queue of the queues and a first counter of the counters, which is associated with said first queue, constitutes the partition management information, and the other is used as the common management information.

9. The storage subsystem according to claim 7, wherein a queue management table is associated with each queue, and the queue management table associated with a queue that constitutes the partition management information is provided in each of the cache partition regions.

10. A storage subsystem according to claim 7, wherein the information in relation to the queue and counter which is to be updated is both queue and counter information in relation to the first cache partition region and queue and counter information in relation to the second cache partition region.

11. A storage subsystem according to claim 7, wherein the information in relation to the queue which is to be updated includes connecting a movement target slot control block table to a movement-destination queue of the second cache partition region.

12. A storage subsystem, comprising:

a plurality of channel adapters that each controls the exchange of data with a host device;

a plurality of storage device groups that each provides a logical storage region;

a plurality of disk adapters that each controls the exchange of data with each of the storage device groups;

a cache memory that is used by each of the channel adapters and each of the disk adapters;

a plurality of cache partition regions constituted by logically partitioning the cache memory; and

a control memory that stores management information for managing each of the cache partition regions,

wherein the management information is constituted by partition management information items corresponding to the cache partition regions, and common management information that is applied to all of the cache partition regions,

wherein the management information includes;

a free queue to which a cache management unit in an unused state is connected and a free queue counter associated with the free queue,

a dirty queue to which a cache management unit for storing data in a dirty state prior to reflection in the storage device group is connected and a dirty queue counter associated with the dirty queue,

a clean queue to which a cache management unit for storing data in a clean state that has been reflected in the storage device group is connected and a clean queue counter associated with the clean queue, and

an in-use counter that counts the total capacity of in-use areas in the cache memory,

wherein the free queue counter, the clean queue, the clean queue counter and the in-use counter are provided in each of the cache partition regions and each constitute the partition management information,

wherein the free queue, the dirty queue, and the dirty queue counter are used as the common management information,

wherein when a cache partition region corresponding to a certain logical storage region is transferred from a first cache partition region to a second cache partition region, information in relation to a queue and counter included in the management information stored in the control memory is updated, and

wherein the information in relation to the counter which is to be undated includes subtracting a count from an in-use counter and a counter in the first cache partition region in accordance with a number of slots moved.

13. A storage subsystem according to claim 12, wherein the information in relation to the queue and counter which is to be updated is both queue and counter information in relation to the first cache partition region and queue and counter information in relation to the second cache partition region.

14. A storage subsystem according to claim 12, wherein the information in relation to the queue which is to be updated includes connecting a movement target slot control block table to a movement-destination queue of the second cache partition region.

15. A storage subsystem, comprising:

a plurality of channel adapters that each controls the exchange of data with a host device;

a plurality of storage device groups that each provides a logical storage region;

a plurality of disk adapters that each controls the exchange of data with each of the storage device groups;

a cache memory that is used by each of the channel adapters and each of the disk adapters;

a plurality of cache partition regions constituted by logically partitioning the cache memory; and

a control memory that stores management information for managing each of the cache partition regions,

wherein the management information is constituted by both a plurality of partition management information that is used to independently manage corresponding ones of the cache partition regions, and common management information that is used to commonly manage a plurality of the cache partition regions,

wherein when a cache partition region corresponding to a certain logical storage region is transferred from a first cache partition region to a second cache partition region, information in relation to a queue and counter included in the management information stored in the control memory is updated, and

wherein the information in relation to the counter which is to be updated includes subtracting a count from an in-use counter and a counter in the first cache partition region in accordance with a number of slots moved.

16. A method for controlling a storage subsystem that comprises:

a plurality of upper interface control units that each controls the exchange of data with a host device,

a plurality of storage device groups that each provides a logical storage region,

a plurality of lower interface control units that each controls the exchange of data with each of the storage device groups, and

a memory section that is used by each of the upper interface control units and each of the lower interface control units,

the method comprising the steps of;

partitioning a cache region provided by the memory section into a plurality of cache partition regions;

wherein management information corresponding to the cache partition regions is provided for independently managing each of the cache partition regions and common management information is provided for commonly managing a plurality of the cache partition regions; and

independently managing data in each of the cache partition regions based on the partition management information and managing data in a plurality of the cache partition region based on the common management information,

wherein when a cache partition region corresponding to a certain logical storage region is transferred from a first cache partition region to a second cache partition region, information in relation to a queue and counter included in the management information is updated, and

wherein the information in relation to the counter which is to be updated includes subtracting a count from an in-use counter and a counter in the first cache partition region in accordance with a number of slots moved.

17. A method according to claim 16, wherein the information in relation to the queue and counter which is to be updated is both queue and counter information in relation to the first cache partition region and queue and counter information in relation to the second cache partition region.

18. A method according to claim 16, wherein the information in relation to the queue which is to be updated includes connecting a movement target slot control block table to a movement-destination queue of the second cache partition region.