System and method for implementing a dynamic cache for a data storage system

• US 7,856,530 B1
• Filed: 10/31/2007
• Issued: 12/21/2010
• Est. Priority Date: 10/31/2007
• Status: Active Grant

• Current Assignee: Network Appliance Incorporated (NetApp, Inc.)
• Original Assignee: Network Appliance Incorporated (NetApp, Inc.)
• Inventors: Mu, Paul Yuedong
• Primary Examiner(s): LANE, JOHN A
• Application Number:
  US11/981,663
• Time in Patent Office: 1,147 Days
• Field of Search: 711/119, 711/145, 711/167
• US Class Current: 711/119
• CPC Class Codes: G06F 12/0866 for peripheral storage systems, e.g. disk cache
• CPC Class Codes: G06F 12/0888 using selective caching, e.g. bypass

Abstract

A dynamic cache system is configured to flexibly respond to changes in operating parameters of a data storage and retrieval system. A cache controller in the system implements a caching policy describing how and what data should be cached. The policy can provide different caching behavior based on exemplary parameters such as a user ID, a specified application or a given workload. The cache controller is coupled to the data path for a data storage system and can be implemented as a filter in a filter framework. The cache memory for storing cached data can be local or remote from the cache controller. The policies implemented in the cache controller permit an application to control caching of data to permit optimization of data flow for the particular application.

Claims

1. A dynamic cache system for use with a data storage system having a data access interface and a storage device with a data path therebetween, the dynamic cache system comprising:

a cache controller coupled to the data path between the storage device and the data access interface, the cache controller registering to be called by a filter framework in response to an I/O operation on the data path between the data access interface and the data storage device to control caching of data accessed through the I/O operation;

a storage memory accessible by the cache controller and configurable by the cache controller to be an intermediate storage for the data accessed through the I/O operation on the data path under control of the cache controller; and

a caching definition accessible by the cache controller having a predetermined criteria for prompting caching activity by the cache controller, the cache controller being responsive to the predetermined criteria to provide selective caching of the data in the storage memory such that the data associated with the I/O operation is cached or prevented from being cached based on the predetermined criteria.

2. The dynamic cache system according to claim 1, further comprising a plurality of caches configured in the storage memory by the cache controller and under control of the cache controller.

3. The dynamic cache system according to claim 2, wherein at least one of the plurality of caches further comprises a read cache or a write cache.

4. The dynamic cache system according to claim 1, wherein the cache controller can allocate and deallocate a cache in the storage memory.

5. The dynamic cache system according to claim 1, further comprising:

a file system implemented in conjunction with the data storage system; and

the predetermined criteria identifies file based parameters in the I/O operation.

6. The dynamic cache system according to claim 1, further comprising:

a logical unit defined in the data storage system accessible for block based I/O operations from the data access interface; and

the predetermined criteria identifies parameters in the block based I/O operation.

7. The dynamic cache system according to claim 1, wherein the predetermined criteria further comprises one or more of a user ID, an application ID, a process ID or a thread ID.

8. The dynamic cache system according to claim 1, wherein the storage memory is remotely located from the cache controller.

9. The dynamic cache system according to claim 1, wherein the cache controller is configured as a filter operable with the filter framework coupled to the data path.

10. The dynamic cache system according to claim 1, further comprising:

the filter framework coupled to the data path in the data storage system for managing filter functions; and

the dynamic cache system being implemented as a filter operable within the filter framework for caching I/O data.

11. The dynamic cache system according to claim 1, wherein the predetermined criteria further comprises one or more of a client source identifier, user credential, file directory, volume name, file offset, time of day, I/O groupings, I/O sequence or process credentials.

12. A method for implementing a dynamic cache system for use with a data storage system having a data access interface and a storage device with a data path therebetween, the method comprising:

configuring a storage memory to be an intermediate storage for data accessed through an I/O operation occurring on the data path between the data access interface and the storage device;

registering to be called by a filter framework in response to the I/O operation being intercepted on the data path between the data access interface and the data storage device;

identifying a parameter within the I/O operation to contribute to determining a caching behavior for the data associated with the I/O operation;

consulting a caching definition determining a caching action for the data to cause or prevent caching of the data based at least in part on a value of the identified parameter; and

caching the data in accordance with the caching action.

13. The method according to claim 12, wherein configuring the storage further comprises allocating a plurality of caches.

14. The method according to claim 13, further comprising implementing at least one cache as a read cache or as a write cache.

15. The method according to claim 12, wherein configuring the storage further comprises allocating or deallocating a cache in the storage memory.

17. The method according to claim 12, further comprising:

defining a block based logical unit in the data storage system;

performing a block based I/O operation on the logical unit; and

identifying a block based parameter within the I/O operation.

18. The method according to claim 12, wherein the caching definition further comprises one or more of a user ID, an application ID, a process ID or a thread ID.

19. The method according to claim 12, further comprising arranging the memory storage at a remote location from the data storage system.

20. The method according to claim 12, further comprising implementing the cache definition in a filter operable with the filter framework implemented in the data storage system.

21. The method according to claim 12, further comprising implementing the filter framework in the data storage system for managing filter framework in the data storage system for managing filter functions;

and

implementing the dynamic cache system as a filter operable within the filter framework for caching I/O data.

22. The method according to claim 12, further comprising providing one or more of a client source identifier, user credential, file directory, volume name, file offset, time of day, I/O groupings, I/O sequence or process credentials in the caching definition.

16. The method according 12, further comprising:

implementing a file system in conjunction with the data storage system; and

identifying a file based parameter within the I/O operation.

23. A data storage and retrieval system having a filter framework operable to provide filtering functions, and a dynamic cache filter being operable with the filter framework and comprising:

a cache controller coupled to the filter framework and operable to send and receive information to and from the filter framework, the cache controller registering to be called by the filter framework based on an I/O operation intercepted by the filter framework to control caching of data accessed through the I/O operation;

a cache memory coupled to the cache controller and configured to be an intermediate storage for the data associated with the I/O under control of the cache controller; and

a caching definition in the cache controller being configured to influence the cache controller to implement caching of the data in accordance with a predetermined criteria such that the data associated with the I/O operation is cached or prevented from being cached based on a parameter associated with the I/O operation in combination with the predetermined criteria.

24. The data storage and retrieval system according to claim 23, further comprising:

a logical unit as a collection of blocks and accessible through a block based I/O operation; and

the caching definition comprises a parameter of the block based I/O operation.

25. The data storage and retrieval system according to claim 23, wherein the parameter comprises one or more of a user ID, application, process, thread, client source identifier, user credential, file directory, volume name, file offset, time of day, I/O grouping, I/O sequence or process credentials.