초록 ▼

A computer-implemented method for relocating deduplicated data within a multi-device storage system. The method may include identifying a set of deduplicated data units stored on a first device of the multi-device storage system. Each data unit in the set of data units is referred to by one or more

A computer-implemented method for relocating deduplicated data within a multi-device storage system. The method may include identifying a set of deduplicated data units stored on a first device of the multi-device storage system. Each data unit in the set of data units is referred to by one or more deduplication references. The method may also include procuring reference data that indicates, for each data unit in the set of deduplicated data units, the number of deduplication references that point to the data unit. The method may further include using the reference data to select one or more data units from the set of deduplicated data units for relocation to a second device in the multi-device storage system and relocating the one or more data units to the second device in the multi-device storage system. Various other methods, systems, and computer-readable media are also disclosed.

대표청구항 ▼

1. A computer-implemented method for relocating deduplicated data within a multi-device storage system, at least a portion of the method being performed by a computing device comprising at least one processor, the method comprising: identifying a set of deduplicated data units stored on a first devi

1. A computer-implemented method for relocating deduplicated data within a multi-device storage system, at least a portion of the method being performed by a computing device comprising at least one processor, the method comprising: identifying a set of deduplicated data units stored on a first device in the multi-device storage system, wherein each data unit in the set of deduplicated data units is referred to by one or more deduplication references;procuring reference data that indicates, for each data unit in the set of deduplicated data units, a number of deduplication references that point to the data unit, wherein the number of references that point to the data unit comprises a count of how many references point to the data unit;using the reference data to select one or more data units from the set of deduplicated data units for relocation to a second device in the multi-device storage system by determining that a number of deduplication references that point to the one or more selected data units is greater than a number of deduplication references that point to one or more other data units in the set of deduplicated data units;relocating the one or more selected data units to the second device in the multi-device storage system by moving the one or more selected data units with relatively high deduplication reference counts to a device with higher performance than the first device. 2. The method of claim 1, wherein the number of deduplication references that point to the data unit includes, for each data unit in the set of deduplicated data units, at least one of: a number of original data objects that point to the data unit;a number of data object revisions that point to the data unit;a number of client devices associated with data objects that point to the data unit. 3. The method of claim 2, further comprising counting at least one of: the number of original data objects that point to the data unit;the number of data object revisions that point to the data unit;the number of client devices associated with data objects that point to the data unit. 4. The method of claim 2, wherein procuring the reference data comprises summing at least two of: the number of original data objects that point to the data unit;the number of data object revisions that point to the data unit;the number of client devices associated with data objects that point to the data unit. 5. The method of claim 4, wherein procuring the reference data comprises: creating a first weighted count by multiplying the number of data object revisions by a first weight;creating a second weighted count by multiplying at least one of the following by a second weight: the number of original data objects that point to the data unit;the number of client devices associated with data objects that point to the data unit;summing the first and second weighted counts to provide the reference count for the data unit, wherein the first weight is greater than the second weight. 6. The method of claim 4, wherein procuring the reference data comprises: creating a first weighted count by multiplying the number of client devices associated with data objects that point to the data unit by a first weight;creating a second weighted count by multiplying at least one of the following by a second weight: the number of original data objects that point to the data unit;the number of data object revisions that point to the data unit;summing the first and second weighted counts to provide the reference count for the data unit, wherein the first weight is greater than the second weight. 7. The method of claim 1, wherein procuring the reference data comprises, for each data unit in the set of deduplicated data units: using historical reference data to select a subset of references that point to the data unit that are more significant than one or more other references that point to the data unit;multiplying references in the subset of references by a higher weight than the other references that point to the data unit. 8. The method of claim 1, wherein procuring the reference data comprises: using historical data unit data to select a data unit from the set of deduplicated data units that is more significant than other data units in the set of deduplicated data units;multiplying the number of deduplication references that point to the selected data unit by a first weight;multiplying the number of deduplication references that point to the other data units in the set of deduplicated data units by a second weight, wherein the first weight is greater than the second weight. 9. The method of claim 1, wherein: the second device in the multi-device storage system comprises a solid-state drive with higher performance for random access than the first device. 10. The method of claim 1, tangibly embodied as computer-executable instructions on at least one computer-readable medium. 11. A system for relocating deduplicated data within a multi-device storage system, the system comprising: at least one processor;an identification module programmed to direct the processor to identify a set of deduplicated data units stored on a first device in the multi-device storage system, wherein each data unit in the set of deduplicated data units is referred to by one or more deduplication references;a data-collection module programmed to direct the processor to procure reference data that indicates, for each data unit in the set of deduplicated data units, a number of deduplication references that point to the data unit, wherein the number of references that point to the data unit comprises a count of how many references point to the data unit;a selection module programmed to direct the processor to use the reference data to select one or more data units from the set of deduplicated data units for relocation to a second device in the multi-device storage system by determining that a number of deduplication references that point to the one or more selected data units is greater than a number of deduplication references that point to one or more other data units in the set of deduplicated data units;a relocation module programmed to direct the processor to relocate the one or more selected data units to the second device in the multi-device storage system by moving the one or more selected data units with relatively high deduplication reference counts to a device with higher performance than the first device. 12. The system of claim 11, wherein the number of deduplication references that point to the data unit includes, for each data unit in the set of deduplicated data units, at least one of: a number of original data objects that point to the data unit;a number of data object revisions that point to the data unit;a number of client devices associated with data objects that point to the data unit. 13. The system of claim 12, wherein the data-collection module is further programmed to direct the processor to count at least one of: the number of original data objects that point to the data unit;the number of data object revisions that point to the data unit;the number of client devices associated with data objects that point to the data unit. 14. The system of claim 12, wherein the data-collection module is programmed to direct the processor to procure the reference count by summing at least two of: the number of original data objects that point to the data unit;the number of data object revisions that point to the data unit;the number of client devices associated with data objects that point to the data unit. 15. The system of claim 14, wherein the data-collection module is programmed to direct the processor to procure the reference count by: creating a first weighted count by multiplying the number of data object revisions by a first weight;creating a second weighted count by multiplying at least one of the following by a second weight: the number of original data objects that point to the data unit;the number of client devices associated with data objects that point to the data unit;summing the first and second weighted counts to provide the reference count for the data unit, wherein the first weight is greater than the second weight. 16. The system of claim 14, wherein the data-collection module is programmed to direct the processor to procure the reference count by: creating a first weighted count by multiplying the number of client devices associated with data objects that point to the data unit by a first weight;creating a second weighted count by multiplying at least one of the following by a second weight: the number of original data objects that point to the data unit;the number of data object revisions that point to the data unit;summing the first and second weighted counts to provide the reference count for the data unit, wherein the first weight is greater than the second weight. 17. The system of claim 11, wherein: the second device in the multi-device storage system comprises a solid-state drive. 18. A computer-implemented method for relocating deduplicated data within a multi-device storage system, at least a portion of the method being performed by a computing device comprising at least one processor, the method comprising: identifying a set of deduplicated data units stored on a first device in the multi-device storage system, wherein each data unit in the set of deduplicated data units is referred to by one or more deduplication references;determining, for each data unit in the set of deduplicated data units, a reference count that indicates a number of deduplication references that point to the data unit, wherein the number of references that point to the data unit comprises a count of how many references point to the data unit;identifying a subset of data units in the set of deduplicated data units by determining that the reference count of each data unit in the subset of data units is higher than the reference count of one or more other data units in the set of deduplicated data units;relocating the subset of data units to a second device in the multi-device storage system. 19. The method of claim 18, wherein: the second device comprises a solid state drive.