System, method and computer program product for deduplication aware quality of service over data tiering

• US 9,298,385 B2
• Filed: 04/12/2013
• Issued: 03/29/2016
• Est. Priority Date: 04/12/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Baldwin, Duane M., Eda, Sasikanth, Ghuge, Deepak R., Olson, John T., Patil, Sandeep R.
• Primary Examiner(s): Thai, Tuan
• Assistant Examiner(s): Sain, Gautam
• Application Number:
  US13/862,270
• Publication Number:
  US 20140310455A1
• Time in Patent Office: 1,082 Days
• Field of Search: 711/154
• US Class Current: 1/1
• CPC Class Codes: G06F 12/00 Accessing, addressing or allocating within memory systems or architectures digital input from, or digital output to record carriers, e.g. to disk storage units, G06F3/06
• CPC Class Codes: G06F 3/0608 Saving storage space on storage systems
• CPC Class Codes: G06F 3/0641 De-duplication techniques
• CPC Class Codes: G06F 3/0649 Lifecycle management
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: G06F 3/0685 Hybrid storage combining heterogeneous device types, e.g. hierarchical storage, hybrid arrays

Abstract

A method for ensuring compliance of service levels corresponding to file-system operations in a deduplicated, tiered storage system comprising storage devices of varying performance levels, each of the performance levels being associated with at least one service level, includes performing a deduplication remapping operation so that a data item corresponding to a higher service level, is stored as an original on a higher performance storage tier while duplicates of the data item corresponding to a lower service level exist in lower performance storage tier at least in part as pointers to the data item on the higher performance storage tier.

Claims

1. A computer program product for ensuring compliance of service levels corresponding to file-system operations in a deduplicated, tiered storage system comprising storage devices of varying performance levels, each of the performance levels being associated with at least one service level, the computer program product comprising a non-transitory computer readable storage medium having program code embodied therewith, the program code readable/executable by a processor to:

perform a deduplication remapping operation, by the processor, so that a data item corresponding to a higher service level is stored as an original data item on a higher performance storage tier while duplicates of the data item corresponding to a lower service level are physically stored in a lower performance storage tier at least in part as pointers to the original data item on the higher performance storage tier;

receive, by the processor, a modification to the original data item on the higher performance storage tier;

store, by the processor, a modified copy of the original data item as a new data item on the higher performance storage tier;

retain, by the processor, the original data item on the higher performance storage tier for a period of time;

maintain, by the processor, the pointers from the duplicates to the original data item for the period of time;

make, by the processor, an entry of the modified copy of the original data item to a relational set;

enable, by the processor, direct saving of the modified copy of the original data item;

transfer, by the processor, the original data item from the higher performance storage tier to the lower performance storage tier after the period of time and at a statistically identified low load time thereby replacing the duplicates on the lower performance storage tier;

delete, by the processor, the original data item from the higher performance storage tier after the transferring;

analyze, by the processor, data item entries and a service level associated with each of the entries; and

perform, by the processor, the deduplication remapping operation to move data items from the lower performance storage tier to the higher performance storage tier when the data items are associated with the higher service level,

wherein the deduplication remapping operation includes;

receiving the data item;

determining a service level associated with the data item;

storing the received data item as an original on the higher performance storage tier in response to determining that the data item corresponds to the higher service level and in response to determining that the data item does not form a duplicate to any data item stored on the lower performance storage tier;

mapping at least a portion of subsequently received duplicate data items to the received data item stored on the higher performance storage tier in response to storing the received data item as an original on the higher performance storage tier, wherein the subsequently received duplicate data items correspond to a lower service level and exist on the lower performance storage tier;

storing the received data item as an original on the lower performance storage tier in response to determining that the data item corresponds to the lower service level and in response to determining that the data item does not form a duplicate to any data item stored on the higher performance storage tier; and

mapping at least a portion of the received data item to subsequently received duplicate data items stored on the higher performance storage tier in response to storing the received data item as an original on the lower performance storage tier, wherein the subsequently received duplicate data items correspond to the higher service level and exist on the higher performance storage tier.

2. The computer program product of claim 1,

wherein the higher service level is associated with a higher subscription cost than the lower service level.

3. The computer program product of claim 1, wherein the computer program product ensures compliance of service levels corresponding to file-system operations in a deduplicated, tiered storage system comprising storage devices of varying performance levels, each of the performance levels being associated with at least one service level, wherein the deduplication remapping operation includes receiving the data item, the data item corresponding to the higher service level, determining that a duplicate of the data item is present in the lower performance storage tier, storing the received data item in the higher performance storage tier, marking the stored data item on the higher performance storage tier as an original, mapping at least a portion of the duplicate data item to the stored data item on the higher performance storage tier.

4. The computer program product of claim 1, wherein the program code is further readable/executable by the processor to receive a modification to the data item on the higher performance storage tier;

store the modified data item as a new data item on the higher performance storage tier;

transfer the data item from the higher performance storage tier to the lower performance storage tier thereby replacing the duplicate on the lower performance storage tier; and

delete the data item from the higher performance storage tier after the transferring.

5. The computer program product of claim 1, wherein the program code is further readable/executable by the processor to detect a request from a user for deletion of the data item from the higher performance storage tier;

represent to the user that the data item has been deleted;

transfer the data item from the higher performance storage tier to the lower performance storage tier thereby replacing the duplicate on the lower performance storage tier; and

delete the data item from the higher performance storage tier after the transferring.

6. The computer program product of claim 1, wherein the program code is further readable/executable by the processor to receive a modification to the duplicate data item on the lower performance storage tier;

store the modification on the lower performance storage tier;

transfer portions of the data item on the higher performance storage tier that correspond to unmodified portions of the duplicate data item to the lower performance storage tier;

store the portions in association with the modification as a new data item.

7. A system, comprising:

a processor and logic integrated with and/or executable by the processor, the logic being configured to;

perform a deduplication remapping operation so that a data item corresponding to a higher service level is stored as an original data item on a higher performance storage tier while duplicates of the data item corresponding to a lower service level are physically stored in a lower performance storage tier at least in part as pointers to the original data item on the higher performance storage tier;

receive, by the processor, a modification to the original data item on the higher performance storage tier;

store, by the processor, a modified copy of the original data item as a new data item on the higher performance storage tier;

retain, by the processor, the original data item on the higher performance storage tier for a period of time;

maintain, by the processor, the pointers from the duplicates to the original data item for the period of time;

make, by the processor, an entry of the modified copy of the original data item to a relational set;

enable, by the processor, direct saving of the modified copy of the original data item;

transfer, by the processor, the original data item from the higher performance storage tier to the lower performance storage tier after the period of time and at a statistically identified low load time thereby replacing the duplicates on the lower performance storage tier;

delete, by the processor, the original data item from the higher performance storage tier after the transferring;

analyze, by the processor, data item entries and a service level associated with each of the entries; and

perform, by the processor, the deduplication remapping operation to move data items from the lower performance storage tier to the higher performance storage tier when the data items are associated with the higher service level,

wherein the deduplication remapping operation includes;

receiving the data item;

determining a service level associated with the data item;

storing the received data item as an original on the higher performance storage tier in response to determining that the data item corresponds to the higher service level and in response to determining that the data item does not form a duplicate to any data item stored on the lower performance storage tier;

mapping at least a portion of subsequently received duplicate data items to the received data item stored on the higher performance storage tier in response to storing the received data item as an original on the higher performance storage tier, wherein the subsequently received duplicate data items correspond to a lower service level and exist on the lower performance storage tier;

storing the received data item as an original on the lower performance storage tier in response to determining that the data item corresponds to the lower service level and in response to determining that the data item does not form a duplicate to any data item stored on the higher performance storage tier; and

mapping at least a portion of the received data item to subsequently received duplicate data items stored on the higher performance storage tier in response to storing the received data item as an original on the lower performance storage tier, wherein the subsequently received duplicate data items correspond to the higher service level and exist on the higher performance storage tier.

8. The system as recited in claim 7, wherein the higher service level is associated with a higher subscription cost than the lower service level.

9. The system as recited in claim 7, wherein the deduplication remapping operation includes receiving the data item, the data item corresponding to the higher service level, determining that a duplicate of the data item is present in the lower performance storage tier, storing the received data item in the higher performance storage tier, marking the stored data item on the higher performance storage tier as an original, mapping at least a portion of the duplicate data item to the stored data item on the higher performance storage tier.

10. The system as recited in claim 7, comprising logic configured to:

detect a request from a user for deletion of the data item from the higher performance storage tier;

represent to the user that the data item has been deleted;

transfer the data item from the higher performance storage tier to the lower performance storage tier thereby replacing the duplicate on the lower performance storage tier; and

delete the data item from the higher performance storage tier after the transferring.

11. The system as recited in claim 7, comprising logic configured to:

receive a modification to the duplicate data item on the lower performance storage tier;

store the modification on the lower performance storage tier;

transfer portions of the data item on the higher performance storage tier that correspond to unmodified portions of the duplicate data item to the lower performance storage tier;

store the portions in association with the modification as a new data item.

12. A computer-implemented method for ensuring compliance of service levels corresponding to file-system operations in a deduplicated, tiered storage system comprising storage devices of varying performance levels, each of the performance levels being associated with at least one service level, the method comprising:

performing, by a computer, a deduplication remapping operation so that a data item corresponding to a higher service level is stored as an original data item on a higher performance storage tier while duplicates of the data item corresponding to a lower service level are physically stored in a lower performance storage tier at least in part as pointers to the original data item on the higher performance storage tier;

receiving, by the computer, a modification to the original data item on the higher performance storage tier;

storing, by the computer, a modified copy of the original data item as a new data item on the higher performance storage tier;

retaining, by the computer, the original data item on the higher performance storage tier for a period of time;

maintaining, by the computer, the pointers from the duplicates to the original data item for the period of time;

making, by the computer, an entry of the modified copy of the original data item to a relational set;

enabling, by the computer, direct saving of the modified copy of the original data item;

transferring, by the computer, the original data item from the higher performance storage tier to the lower performance storage tier after the period of time and at a statistically identified low load time thereby replacing the duplicates on the lower performance storage tier;

deleting, by the computer, the original data item from the higher performance storage tier after the transferring;

analyzing, by the computer, data item entries and a service level associated with each of the entries; and

performing, by the computer, the deduplication remapping operation to move data items from the lower performance storage tier to the higher performance storage tier when the data items are associated with the higher service level,

wherein the deduplication remapping operation includes;

receiving the data item;

determining a service level associated with the data item;

storing the received data item as an original on the higher performance storage tier in response to determining that the data item corresponds to the higher service level and in response to determining that the data item does not form a duplicate to any data item stored on the lower performance storage tier;

mapping at least a portion of subsequently received duplicate data items to the received data item stored on the higher performance storage tier in response to storing the received data item as an original on the higher performance storage tier, wherein the subsequently received duplicate data items correspond to a lower service level and exist on the lower performance storage tier;

storing the received data item as an original on the lower performance storage tier in response to determining that the data item corresponds to the lower service level and in response to determining that the data item does not form a duplicate to any data item stored on the higher performance storage tier; and

mapping at least a portion of the received data item to subsequently received duplicate data items stored on the higher performance storage tier in response to storing the received data item as an original on the lower performance storage tier, wherein the subsequently received duplicate data items correspond to the higher service level and exist on the higher performance storage tier.

13. The method as recited in claim 12, with a proviso that a data item corresponding to the higher service level is not stored as a reference to a duplicate data item on the lower performance storage tier, wherein the higher service level is associated with a higher subscription cost than the lower service level.

14. The method as recited in claim 12, wherein the deduplication remapping operation includes receiving the data item, the data item corresponding to the higher service level, determining that a duplicate of the data item is present in the lower performance storage tier, storing the received data item in the higher performance storage tier, marking the stored data item on the higher performance storage tier as an original, mapping at least a portion of the duplicate data item to the stored data item on the higher performance storage tier using the pointers.

15. The method as recited in claim 12, comprising receiving a modification to the data item on the higher performance storage tier;

storing the modified data item as a new data item on the higher performance storage tier;

transferring the data item from the higher performance storage tier to the lower performance storage tier thereby replacing the duplicate on the lower performance storage tier; and

deleting the data item from the higher performance storage tier after the transferring.

16. The method as recited in claim 12, comprising detecting a request from a user for deletion of the data item from the higher performance storage tier;

representing to the user that the data item has been deleted;

transferring the data item from the higher performance storage tier to the lower performance storage tier thereby replacing the duplicate on the lower performance storage tier; and

deleting the data item from the higher performance storage tier after the transferring.

17. The method as recited in claim 12, comprising receiving a modification to the duplicate data item on the lower performance storage tier;

storing the modification on the lower performance storage tier;

transferring portions of the data item on the higher performance storage tier that correspond to unmodified portions of the duplicate data item to the lower performance storage tier;

storing the portions in association with the modification as a new data item.