STORAGE SYSTEM USING A RAPID STORAGE DEVICE AS A CACHE

• US 20110302365A1
• Filed: 10/30/2009
• Published: 12/08/2011
• Est. Priority Date: 02/13/2009
• Status: Abandoned Application

• Current Assignee: OCZ Storage Solutions Incorporated (Toshiba Corporation)
• Original Assignee: Indilinx Co. Ltd. (ZCO Liquidating Corporation)
• Inventors: Park, Han-Mook, Heo, Jin Yong, An, Hyung Geun
• Application Number:
  US13/201,362
• Publication Number:
  US 20110302365A1
• Time in Patent Office: Days
• US Class Current: 711/104
• CPC Class Codes: G06F 12/0866 for peripheral storage systems, e.g. disk cache

Abstract

Provided is a storage system using a high speed storage device as a cache. The storage system includes a large-volume of first storage device, a high speed second storage device, and a Random Access Memory (RAM). The large-volume of first storage device corresponds to a Hard Disk Drive (HDD), and the high speed second storage device corresponds to a Solid State Drive (SSD). Also, the high speed second drive is used as a cache. The first storage device manages content files super block by super block, and the second storage device manages cache files block by block.

Claims

1. A storage system using a high speed storage device as a cache, the system comprising:

a first storage device to store a plurality of content files including a plurality of blocks, and to manage the plurality of blocks in super block units, the super block having a predetermined size;

a second storage device to cache at least one super block among the plurality of super blocks, and to manage the at least one super block in block units; and

a Random Access Memory (RAM) to store at least one block among the plurality of blocks,

wherein a capacity of the first storage device is greater than a capacity of the second storage device, or an access rate of the second storage device is higher than an access rate of the first storage device.

2. The system of claim 1, further comprising:

an input/output scheduler to check whether a target block is stored in the second storage device or the RAM in response to a read request from an application related to the target block, and to return the target block extracted from at least one of the first storage device, the second storage device, and the RAM based on a result of the check.

3. The system of claim 2, wherein the input/output scheduler uses a high speed storage device that manages the first storage device and the second storage device as a cache so that the application may recognize the first storage device and the second storage device as a single logical storage device.

4. The system of claim 2, further comprising:

a cache file controller to read the target block from the second storage device using a cache file block mapping table where location information about a plurality of blocks included in at least one super block cached by the second storage device is recorded.

5. The system of claim 2, wherein the input/output scheduler uses a content super block mapping table comprising at least one of information regarding whether the plurality of super blocks is cached by the second storage device, location information regarding the at least one super block that is cached by the second storage device, information regarding whether the plurality of super blocks is stored in the RAM, information regarding an access time of the plurality of super blocks or a number of times that the plurality of super blocks is accessed, and information regarding whether the plurality of super blocks is currently being used.

6. The system of claim 1, wherein the second storage device caches the at least one super block based on a number of times that the plurality of super blocks is accessed, or a number of times that the plurality of content files is accessed.

7. The system of claim 1, wherein, when a plurality of user sessions exists,

the second storage device caches a super block used in a predetermined user session based on a super block used in at least one remaining user session.

8. The system of claim 7, wherein the second storage device caches the super block used in the predetermined user session depending on whether at least one super block adjacent to the super block used in the predetermined user session is used in the at least one remaining user session.

9. The system of claim 1, wherein the first storage device returns a super block including the target block among the at least one super block when a read request from an application related to the target block is received.

10. The system of claim 1, wherein the first storage device corresponds to a Hard Disk Drive (HDD), and the second storage device corresponds to a Solid State Drive (SSD).

11. The system of claim 1, wherein the storage system is used for a server streaming multimedia contents to a plurality of users.

12. The system of claim 4, wherein, when a super block to be deleted, or a block to be deleted exists among the at least one super block cached by the second storage device,

the input/output scheduler updates the cache file block mapping table instead of deleting the super block to be deleted, or the block to be deleted, from the second storage device.

13. A reading method of a storage system using a high speed storage device as a cache, the reading method comprising:

receiving a read request from an application related to a target block;

checking whether the target block is stored in a second storage device or a Random Access Memory (RAM); and

returning the target block extracted from at least one of a first storage device, the second storage device, and the RAM based on a result of the check,

when a plurality of content files including a plurality of blocks is stored in the first storage device, the first storage device manages the plurality of blocks in super block units, the second storage device caches at least one super block among the plurality of super blocks, and the RAM stores at least one block among the plurality of blocks, the super block having a predetermined size.

14. The reading method of claim 13, wherein the returning comprises reading the target block from the second storage device using a cache file block mapping table where location information about a plurality of blocks included in at least one super block cached by the second storage device is recorded.

15. The reading method of claim 13, wherein the checking comprises checking whether the target block is stored in the second storage device or the RAM using a content super block mapping table including information regarding whether the plurality of super blocks is cached by the second storage device, and information regarding whether the plurality of super blocks is stored in the RAM.

16. The reading method of claim 13, further comprising:

processing the target block or a super block including the target block to be cached by the second storage device when the target block is stored in the RAM.

20. A computer readable recording medium storing a program to cause a computer to implement the method of one of claim 13.

17. A caching method of a storage system using a high speed storage device as a cache, the caching method comprising:

storing a target block or a super block, among at least one block stored in a Random Access Memory (RAM), including the target block in a second storage device, based on a number of times that the super block including the target block is accessed or a number of times that a content file including the super block is accessed; and

storing the target block or the super block including the target block in the second storage device, depending on whether at least one super block adjacent to the super block including the target block used in a predetermined user session is used in at least one remaining user session when a plurality of user sessions exists,

when a plurality of content files including a plurality of blocks is stored in a first storage device, the first storage device manages the plurality of blocks in super block units, the second storage devices caches at least one super block among the plurality of super blocks, and the RAM stores at least one block among the plurality of blocks, the super block having a predetermined size.

18. The caching method of claim 17, further comprising:

releasing the target block or the super block including the target block from the RAM when the target block or the super block including the target block is not stored in the second storage device.

19. The caching method of claim 18, wherein the storing of the target block or of the super block, among the at least one block stored in the RAM, including the target block in the second storage device, based on the number of times that the super block including the target block is accessed, or the number of times that the content file including the super block is accessed comprises comparing, with a predetermined threshold value, the number of times that the super block including the target block is accessed or the number of times that the content file including the super block is accessed.