Pattern matching using hash tables in storage system
A system and method for managing tables in a storage system is described.
1. A system comprising:
a storage device;
a first fingerprint table and a second fingerprint table, the first table configured to store entries corresponding to data stored in the storage device which has been deemed to be more likely to be successfully deduplicated than other data in the storage device, the second fingerprint table configured to store entries corresponding to data stored in the storage device which has been deemed to be less likely to be successfully deduplicated than other data in the storage device.
2. The system as recited in claim 1 , further comprising:
a storage controller coupled to the storage device, the storage controller configured to:
maintain attributes corresponding to usage of data objects storage in the storage device;
read the attributes corresponding to the objects, responsive to detecting an event, and modify one of the first fingerprint table or the second fingerprint table.
3. The system as recited in claim 2 , wherein the attributes comprise at least one of, access, data age, device performance, device heath, error correction use data, deduplication rate, or read shift voltage.
4. The system as recited in claim 2 , wherein the event comprises one of a garbage collection operation, a health binning operation, a block calibration operation, a device access operation and a wear levelling operation.
5. The system as recited in claim 1 , wherein the data includes patterns.
6. The system as recited in claim 1 , wherein the system is a flash system.
7. The system as recited in claim 1 , wherein the entries comprise hashes.
8. A method comprising:
selecting a first table for storage of an entry corresponding to data stored in a data storage medium, responsive to determining the data being more likely to be successfully deduplicated than other data in the storage device; and
selecting a second table for storage of an entry corresponding to data stored in a data storage medium, responsive to determining the data being less likely to be successfully deduplicated than other data in the storage device.
9. The method of claim 8 , wherein a storage controller is coupled to the storage device, the storage controller configured to:
maintain attributes corresponding to usage of data objects storage in the storage device;
read the attributes corresponding to the objects, responsive to detecting an event, and
modify one of the first fingerprint table or the second fingerprint table.
10. The method of claim 9 , wherein the attributes comprise at least one of, access, data age, device performance, device heath, error correction use data, deduplication rate, or read shift voltage.
11. The method of claim 9 , wherein the event comprises one of a garbage collection operation, a health binning operation, a block calibration operation, a device access operation and a wear levelling operation.
12. The method of claim 8 , wherein the data includes patterns.
13. The method of claim 8 , wherein the first entry comprises a hash.
14. A non-transitory computer readable storage medium comprising program instructions, wherein said program instructions are executable to:
selecting a first table for storage of an entry corresponding to data stored in a data storage medium, responsive to determining the data being more likely to be successfully deduplicated than other data in the storage device; and
selecting a second table for storage of an entry corresponding to data stored in a data storage medium, responsive to determining the data being less likely to be successfully deduplicated than other data in the storage device.
15. The non-transitory computer readable storage medium of claim 14 , wherein a storage controller is coupled to the storage device, the storage controller configured to:
maintain attributes corresponding to usage of data objects storage in the storage device;
read the attributes corresponding to the objects, responsive to detecting an event, and
modify one of the first fingerprint table or the second fingerprint table.
16. The non-transitory computer readable storage medium of claim 15 , wherein the attributes comprise at least one of, access, data age, device performance, device heath, error correction use data, deduplication rate, or read shift voltage.
17. The non-transitory computer readable storage medium of claim 15 , wherein the event comprises one of a garbage collection operation, a health binning operation, a block calibration operation, a device access operation and a wear levelling operation.
18. The non-transitory computer readable storage medium of claim 14 , wherein the data includes patterns.
19. The non-transitory computer readable storage medium of claim 14 , wherein the entry comprises a hash.
20. The non-transitory computer readable storage medium of claim 14 , wherein the data storage medium is flash storage.