Machine learning-based data object storage

• US 10,592,145 B2
• Filed: 02/14/2018
• Issued: 03/17/2020
• Est. Priority Date: 02/14/2018
• Status: Active Grant

• Current Assignee: CommVault Systems Incorporated
• Original Assignee: CommVault Systems Incorporated
• Inventors: Bedadala, Pavan Kumar Reddy, Veeramachaneni, Praveen
• Primary Examiner(s): Ruiz, Aracelis
• Application Number:
  US15/896,943
• Publication Number:
  US 20190250839A1
• Time in Patent Office: 762 Days
• CPC Class Codes: G06F 11/1446 Point-in-time backing up or restoration of persistent data
• CPC Class Codes: G06F 11/1451 by selection of backup contents
• CPC Class Codes: G06F 11/1453 using de-duplication of the data
• CPC Class Codes: G06F 11/1458 Management of the backup or restore process
• CPC Class Codes: G06F 11/1461 Backup scheduling policy
• CPC Class Codes: G06F 11/1469 Backup restoration techniques
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence
• CPC Class Codes: G06F 11/2097 maintaining the standby controller/processing unit updated initialisation or re-synchronisation thereof G06F11/1658 and subgroups
• CPC Class Codes: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: G06F 2201/83 the solution involving signatures
• CPC Class Codes: G06F 2201/84 Using snapshots, i.e. a logical point-in-time copy of the data
• CPC Class Codes: G06F 3/0605 by facilitating the interaction with a user or administrator
• CPC Class Codes: G06F 3/0614 Improving the reliability of storage systems
• CPC Class Codes: G06F 3/0638 Organizing or formatting or addressing of data
• CPC Class Codes: G06F 3/065 Replication mechanisms
• CPC Class Codes: G06F 3/0653 Monitoring storage devices or systems
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: G06N 20/00 Machine learning

Abstract

An information management system is provided herein that uses machine learning (ML) to predict what data to store in a secondary storage device and/or when to perform the storage. For example, a client computing device can be initially configured to store data in a secondary storage device according to one or more storage policies. A media agent in the information management system can monitor data usage on the client computing device, using the data usage data to train a data storage ML model. The data storage ML model may be trained such that the model predicts what data to store in a secondary storage device and/or when to perform the storage. The client computing device can then be configured to use the trained data storage ML model in place of the storage polic(ies) to determine which data to store in a secondary storage device and/or when to perform the storage.

Claims

1. A networked information management system comprising:

a client computing device having one or more first hardware processors, wherein the client computing device is configured with first computer-executable instructions that, when executed, cause the client computing device to store one or more data objects in a secondary storage device according to a storage policy at a first time; and

one or more computing devices in communication with the client computing device, wherein the one or more computing devices each have one or more second hardware processors, wherein the one or more computing devices are configured with second computer-executable instructions that, when executed, cause the one or more computing devices to;

retrieve data object usage data associated with the client computing device;

train a data storage machine learning (ML) model using the data object usage data; and

transmit the data storage ML model to the client computing device such that the client computing device uses the data storage ML model instead of the storage policy to determine which of the one or more data objects to store in the secondary storage device at a second time after the first time.

2. The networked information management system of claim 1, wherein the second computer-executable instructions, when executed, further cause the one or more computing devices to:

retrieve user directory information from a user directory system, wherein the user directory information comprises an indication of an active or inactive status of one or more user credentials; and

train the data storage ML model using the data object usage data and the user directory information.

3. The networked information management system of claim 1, wherein the second computer-executable instructions, when executed, further cause the one or more computing devices to:

retrieve second data object usage data associated with the client computing device;

retrieve the data storage ML model;

retrain the data storage ML model using the second data object usage data; and

transmit the retrained data storage ML model to the client computing device such that the client computing device uses the retrained data storage ML model instead of the data storage ML model to determine which of the one or more data objects to store in the secondary storage device at a third time after the second time.

4. The networked information management system of claim 1, wherein the second computer-executable instructions, when executed, further cause the one or more computing devices to:

obtain a request to store in the secondary storage device a first data object of the one or more data objects from the client computing device, wherein the client computing device generates the request in response to a prediction produced by the data storage ML model;

process the first data object to form a secondary copy of the first data object; and

store the secondary copy of the first data object in the secondary storage device.

5. The networked information management system of claim 1, wherein the data storage ML model generates a prediction, with an associated confidence level, identifying a first data object of the one or more data objects to store in the secondary storage device at the second time after the first time in response to one or more inputs.

6. The networked information management system of claim 5, wherein the one or more inputs comprise at least one of a current time, an identification of a second data object of the one or more data objects generated by a first application running on the client computing device, an age of the second data object, a name of the second data object, a size of the second data object, a data object type of the second data object, or information identifying an active or inactive status of one or more user credentials.

7. The networked information management system of claim 1, wherein the second computer-executable instructions, when executed, further cause the one or more computing devices to train the data storage ML model by deriving patterns from the data object usage data.

8. The networked information management system of claim 1, wherein the data object usage data comprises at least one of data object access times, data object permissions, data object ownership information, data object datapath information, information indicating which application running on the client computing device generated a data object, data object size, data object type, or data object name information.

9. The networked information management system of claim 1, wherein the one or more data objects comprise at least one of a file, a folder, a directory, a file system volume, a data block, or an extent.

10. A computer-implemented method comprising:

retrieving data object usage data associated with a client computing device, the client computing device having one or more first hardware processors, wherein the client computing device is configured with computer-executable instructions that, when executed, cause the client computing device to store one or more data objects in a secondary storage device according to a storage policy at a first time;

training a data storage machine learning (ML) model using the data object usage data; and

transmitting the data storage ML model to the client computing device such that the client computing device uses the data storage ML model instead of the storage policy to determine which of the one or more data objects to store in the secondary storage device at a second time after the first time.

11. The computer-implemented method of claim 10, wherein training a data storage ML model further comprises:

retrieving user directory information from a user directory system, wherein the user directory information comprises an indication of an active or inactive status of one or more user credentials; and

training the data storage ML model using the data object usage data and the user directory information.

12. The computer-implemented method of claim 10, further comprising:

retrieving second data object usage data associated with the client computing device;

retrieving the data storage ML model;

retraining the data storage ML model using the second data object usage data; and

transmitting the retrained data storage ML model to the client computing device such that the client computing device uses the retrained data storage ML model instead of the data storage ML model to determine which of the one or more data objects to store in the secondary storage device at a third time after the second time.

13. The computer-implemented method of claim 10, further comprising:

receiving a request to store in the secondary storage device a first data object of the one or more data objects from the client computing device, wherein the client computing device generates the request in response to a prediction produced by the data storage ML model;

processing the first data object to form a secondary copy of the first data object; and

storing the secondary copy of the first data object in the secondary storage device.

14. The computer-implemented method of claim 10, wherein the data storage ML model generates a prediction, with an associated confidence level, identifying a first data object of the one or more data objects to store in the secondary storage device at the second time after the first time in response to one or more inputs.

15. The computer-implemented method of claim 14, wherein the one or more inputs comprise at least one of a current time, an identification of a second data object of the one or more data objects generated by a first application running on the client computing device, an age of the second data object, a name of the second data object, a size of the second data object, a data object type of the second data object, or information identifying an active or inactive status of one or more user credentials.

16. The computer-implemented method of claim 10, wherein training a data storage ML model further comprises training the data storage ML model by deriving patterns from the data object usage data.

17. The computer-implemented method of claim 10, wherein the data object usage data comprises at least one of data object access times, data object permissions, data object ownership information, data object datapath information, information indicating which application running on the client computing device generated a data object, data object size, data object type, or data object name information.

18. The computer-implemented method of claim 10, wherein the one or more data objects comprise at least one of a file, a folder, a directory, a file system volume, a data block, or an extent.

19. A networked information management system comprising:

a client computing device having one or more first hardware processors, wherein the client computing device is configured with first computer-executable instructions that, when executed, cause the client computing device to store one or more data objects in a secondary storage device according to a storage policy at a first time; and

one or more computing devices in communication with the client computing device, wherein the one or more computing devices each have one or more second hardware processors, wherein the one or more computing devices are configured with second computer-executable instructions that, when executed, cause the one or more computing devices to;

retrieve data object usage data associated with the client computing device;

train a data storage machine learning (ML) model using the data object in response to a threshold amount of the data object usage being retrieved; and

transmit the data storage ML model to the client computing device such that the client computing device uses at least one of the data storage ML model or the storage policy to determine which of the one or more data objects to store in the secondary storage device at a second time after the first time.

20. The networked information management system of claim 19, wherein the second computer-executable instructions, when executed, further cause the one or more computing devices to:

retrieve second data object usage data associated with the client computing device;

retrieve the data storage ML model;

retrain the data storage ML model using the second data object usage data; and

transmit the retrained data storage ML model to the client computing device such that the client computing device uses at least one of the retrained data storage ML model or the storage policy instead of the data storage ML model to determine which of the one or more data objects to store in the secondary storage device at a third time after the second time.