US2016085630A1PendingUtilityA1

Hash collision recovery in a deduplication vault

47
Assignee: STORAGECRAFT TECHNOLOGY CORPPriority: Sep 22, 2014Filed: Jun 24, 2015Published: Mar 24, 2016
Est. expirySep 22, 2034(~8.2 yrs left)· nominal 20-yr term from priority
G06F 11/1076G06F 11/1402G06F 17/30097G06F 16/128G06F 2211/1028G06F 11/1453G06F 16/137G06F 16/90
47
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

Hash collision recovery in a deduplication vault. In one example embodiment, a method for hash collision recovery in a deduplication vault includes creating first parity data for all unique blocks of a source storage at a point in time. The first parity data includes both the unique blocks as well as an order of block positions of the unique blocks as stored in the source storage. Next, a hash value is generated for each of the unique blocks. Then, a backup is stored in a deduplication vault including each of the unique blocks together with its corresponding hash value. Next, second parity data is created for all of the unique blocks of the backup. Then, the first parity data is compared to the second parity data to determine whether one or more hash collisions occurred resulting in one or more missing unique blocks. Next, responsive to the one or more hash collisions occurring, the first parity data is used to recover the one or more missing unique blocks. Then, the backup is restored.

Claims

exact text as granted — not AI-modified
1 . One or more non-transitory computer-readable media storing one or more programs that causes one or more processors to execute a method for hash collision recovery in a deduplication vault, the method comprising:
 creating first parity data for all unique blocks of a source storage at a point in time, the first parity data including both the unique blocks as well as an order of block positions of the unique blocks as stored in the source storage;   generating a hash value for each of the unique blocks;   storing a backup in the deduplication vault including each of the unique blocks together with its corresponding hash value;   creating second parity data for all of the unique blocks of the backup;   comparing the first parity data and the second parity data to determine whether one or more hash collisions occurred resulting in one or more missing unique blocks;   responsive to the one or more hash collisions occurring, using the first parity data to recover the one or more missing unique blocks; and   restoring the backup.

Cited by (0)

No later patents cite this yet.

References (0)

No backward citations on record.