Method for lineage sampling to efficiently detect corruptions

Patent metadata

Field	Value
Publication number	US-2024403286-A1
Application number	US-202418800404-A
Country	US
Kind code	A1
Filing date	Aug 12, 2024
Priority date	Sep 30, 2022
Publication date	Dec 5, 2024
Grant date	—

Abstract

Corruption detection in backups is disclosed. Backups that are received into a backup environment are stored in corresponding lineages. A detection engine is configured to perform corruption detection operations on the most recent backups in each of the lineages based on a sample frequency. Corruption detection operations may also be performed randomly and based on unexpected or unusual changes in backup metadata.

First claim

1 . A method for performing a corruption detection operation that includes read operations in a data protection system, comprising:
receiving backups into the data protection system configured to store the backups, wherein the backups are associated with corresponding lineages, wherein each of the lineages is associated with a sample frequency that is independent of a backup cadence; selecting lineages subject to the corruption detection operation based on the sample frequencies of the lineages and based on a read capacity associated with the data protection system; and performing the corruption detection operation on a most recent backup in each of the selected lineages according to the associated sample frequency.

2 . The method of claim 1 , further comprising pruning at least one of the selected linages when the read capacity is above a threshold.

3 . The method of claim 2 , wherein pruning includes skipping a lineage with a comparatively lower sample frequency or skipping a lineage stored in a lower tier of storage.

4 . The method of claim 1 , wherein the backups comprise synthetic full backups, incremental backups, and/or always full backups and wherein the corruption detection operation is not performed on some of the backups in some of the lineages.

5 . The method of claim 1 , wherein the corruption detection operation includes generating statistics, inferences, and/or probabilities related to malware or other unwanted content.

6 . The method of claim 1 , wherein the sample frequency is included in metadata associated with the backups.

7 . The method of claim 1 , further comprising selecting at least one of the lineages randomly rather than the associated sample frequency.

8 . The method of claim 1 , further comprising triggering dynamic scans based on metadata changes.

9 . The method of claim 8 , wherein metadata changes that trigger dynamic scans include a file size change greater than a threshold amount or an unexpected input/output (IO) pattern.

10 . The method of claim 1 , wherein the backup environment comprises a physical or virtual appliance that is accessed via an air gap.

11 . The method of claim 1 , further comprising recovering from a copy that is clean.

12 . A non-transitory storage medium having stored therein instructions that are executable by one or more hardware processors to perform operations for performing a corruption detection operation that includes read operations in a data protection system, the operations comprising:
receiving backups into the data protection system configured to store the backups, wherein the backups are associated with corresponding lineages, wherein each of the lineages is associated with a sample frequency that is independent of a backup cadence; selecting lineages subject to the corruption detection operation based on the sample frequencies of the lineages and based on a read capacity associated with the data protection system; and performing the corruption detection operation on a most recent backup in each of the selected lineages according to the associated sample frequency.

13 . The method of claim 1 , further comprising pruning at least one of the selected linages when the read capacity is above a threshold.

14 . The method of claim 2 , wherein pruning includes skipping a lineage with a comparatively lower sample frequency or skipping a lineage stored in a lower tier of storage.

15 . The method of claim 1 , wherein the backups comprise synthetic full backups, incremental backups, and/or always full backups and wherein the corruption detection operation is not performed on some of the backups in some of the lineages.

16 . The method of claim 1 , wherein the corruption detection operation includes generating statistics, inferences, and/or probabilities related to malware or other unwanted content.

17 . The method of claim 1 , wherein the sample frequency is included in metadata associated with the backups.

18 . The method of claim 1 , further comprising selecting at least one of the lineages randomly rather than the associated sample frequency.

19 . The method of claim 1 , further comprising triggering dynamic scans based on metadata changes, wherein metadata changes that trigger dynamic scans include a file size change greater than a threshold amount or an unexpected input/output (IO) pattern.

20 . The method of claim 1 , wherein the backup environment comprises a physical or virtual appliance that is accessed via an air gap, further comprising recovering from a copy that is clean.

Assignees

• Dell Products Lp

Inventors

• Madan Nitin
• Bhanjois Bhimsen
• Gowda Kiran Kumar Malle

Classifications

• G06F11/1451 Primary

  by selection of backup contents · CPC title

• G06F2201/84

  Using snapshots, i.e. a logical point-in-time copy of the data · CPC title

• G06F16/2365 Primary

  Ensuring data consistency and integrity · CPC title