Context-driven data backup and recovery
Disclosed herein are systems, methods, and processes to perform context-driven (or context-based) data backup and recovery operations. A request to perform a backup operation on a dataset is received. Current external context datasets related to the dataset and generated based on prioritization techniques are collected from computing devices. a saved context dataset is generated based on the current external context datasets. The backup operation is performed by storing a backup image that includes at least a portion of the dataset and the saved context dataset.
1. A method of comprising:
receiving a request to perform a backup operation on a first dataset, wherein the request identifies a topic;
generating a current external context dataset, wherein
a preliminary identification of a subset of the first dataset is based on an identification of at least one keyword in the subset that is related to the topic, and
generating the current external context dataset comprises applying at least a first prioritization technique to the subset, and
the first prioritization technique is a topic modeling technique, wherein
the topic modeling technique is based, at least in part, on a natural language processing (NLP) methodology, and
the NLP methodology is configured to determine whether the subset is responsive to the topic, at least in part, by applying a lexical analysis technique to the subset;
generating a saved context dataset, wherein
the saved context dataset is generated by determining one or more differences between a previous saved context dataset and the current external context dataset; and
performing the backup operation, wherein
the backup operation comprises storing a backup image, wherein
the backup image comprises
at least a portion of the first dataset, and
the saved context dataset.
2. The method of claim 1 , further comprising:
identifying one or more previous saved context datasets, wherein
the one or more previous saved context datasets are stored temporally and incrementally as part of one or more previous backup images, and
the one or more previous saved context datasets are associated with one or more previous backed up datasets; and
generating the saved context dataset is further based on a difference between the one or more previous saved context datasets and the current external context dataset.
3. The method of claim 1 , wherein
performing the backup operation further comprises storing backup metadata associated with the first dataset as part of the backup image.
4. The method of claim 1 , wherein
generating the current external context dataset is further based, at least in part, on at least a second prioritization technique, wherein
the second prioritization technique is a social network data analysis technique, and
the social network data analysis technique comprises analyzing data associated with one or more users who share a social network with a given user.
5. The method of claim 2 , further comprising:
generating a temporal index based on the saved context dataset and the one or more previous saved context datasets, wherein
the temporal index permits the saved context dataset and the one or more previous saved context datasets to be searched temporally.
6. The method of claim 5 , wherein
each of the one or more previous saved context datasets is associated with at least one previous backup image of the one or more previous backup images; and
the backup image is stored incrementally along with the one or more previous backup images.
7. The method of claim 5 , wherein
the temporal index maintains mapping information between the saved context dataset and the one or more previous saved context datasets, and the backup image and the one or more previous backup images.
8. The method of claim 1 , wherein
generating the current external context dataset is further based, at least in part, on at least one additional prioritization technique, wherein
the at least one additional prioritization technique is at least one of
a cluster analysis technique, or
a graph analysis technique.
9. A non-transitory computer readable storage medium comprising program instructions executable to perform a method comprising:
receiving a request to perform a backup operation on a first dataset, wherein
the request identifies a topic;
generating a current external context dataset, wherein
a preliminary identification of a subset of the first dataset is based on an identification of at least one keyword in the subset that is related to the topic, and
generating the current external context dataset comprises applying at least a first prioritization technique to the subset, and
the first prioritization technique is a topic modeling technique, wherein
the topic modeling technique is based, at least in part, on a natural language processing (NLP) methodology, and
the NLP methodology is configured to determine whether the subset is responsive to the topic, at least in part, by applying a lexical analysis technique to the subset;
generating a saved context dataset, wherein
the saved context dataset is generated by determining one or more differences between a previous saved context dataset and the current external context dataset; and
performing the backup operation, wherein
the backup operation comprises storing a backup image, wherein
the backup image comprises
at least a portion of the first dataset, and
the saved context dataset.
10. The non-transitory computer readable storage medium of claim 9 , wherein the method further comprises:
identifying one or more previous saved context datasets, wherein
the one or more previous saved context datasets are stored temporally and incrementally as part of one or more previous backup images, and
the one or more previous saved context datasets are associated with one or more previous backed up datasets; and
generating the saved context dataset is further based on a difference between the one or more previous saved context datasets and the current external context dataset.
11. The non-transitory computer readable storage medium of claim 9 , wherein
performing the backup operation further comprises storing backup metadata associated with the first dataset as part of the backup image.
12. The non-transitory computer readable storage medium of claim 9 , wherein
generating the current external context dataset is further based, at least in part, on at least a second prioritization technique, wherein
the second prioritization technique is a social network data analysis technique, and
the social network data analysis technique comprises analyzing data associated with one or more users who share a social network with a given user.
13. The non-transitory computer readable storage medium of claim 10 , wherein the method further comprises:
generating a temporal index based on the saved context dataset and the one or more previous saved context datasets, wherein
the temporal index permits the saved context dataset and the one or more previous saved context datasets to be searched temporally.
14. The non-transitory computer readable storage medium of claim 13 , wherein
each of the one or more previous saved context datasets is associated with at least one previous backup image of the one or more previous backup images; and
the backup image is stored incrementally along with the one or more previous backup images.
15. The non-transitory computer readable storage medium of claim 13 , wherein
the temporal index maintains mapping information between the saved context dataset and the one or more previous saved context datasets, and the backup image and the one or more previous backup images.
16. A system comprising:
one or more processors; and
a memory coupled to the one or more processors, wherein the memory stores program instructions executable by the one or more processors to perform a method comprising:
receiving a request to perform a backup operation on a first dataset, wherein
the request identifies a topic;
generating a current external context dataset, wherein
a preliminary identification of a subset of the first dataset is based on an identification of at least one keyword in the subset that is related to the topic, and
generating the current external context dataset comprises applying at least a first prioritization technique to the subset, and
the first prioritization technique is a topic modeling technique, wherein
the topic modeling technique is based, at least in part, on a natural language processing (NLP) methodology, and
the NLP methodology is configured to determine whether the subset is responsive to the topic, at least in part, by applying a lexical analysis technique to the subset;
generating a saved context dataset, wherein
the saved context dataset is generated by determining one or more differences between a previous saved context dataset and the current external context dataset; and
performing the backup operation, wherein
the backup operation comprises storing a backup image, wherein
the backup image comprises
at least a portion of the first dataset, and
the saved context dataset.
17. The system of claim 16 , wherein the method further comprises:
identifying one or more previous saved context datasets, wherein
the one or more previous saved context datasets are stored temporally and incrementally as part of one or more previous backup images, and
the one or more previous saved context datasets are associated with one or more previous backed up datasets; and
generating the saved context dataset is further based on a difference between the one or more previous saved context datasets and the current external context dataset.
18. The system of claim 16 , wherein
performing the backup operation further comprises storing backup metadata associated with the first dataset as part of the backup image.
19. The system of claim 16 , wherein
generating the current external context dataset is further based, at least in part, on at least a second prioritization technique, wherein
the second prioritization technique is a social network data analysis technique, and
the social network data analysis technique comprises analyzing data associated with one or more users who share a social network with a given user.
20. The system of claim 17 , wherein the method further comprises:
generating a temporal index based on the saved context dataset and the one or more previous saved context datasets, wherein
the temporal index permits the saved context dataset and the one or more previous saved context datasets to be searched temporally.
21. The system of claim 20 , wherein
each of the one or more previous saved context datasets is associated with at least one previous backup image of the one or more previous backup images;
the backup image is stored incrementally along with the one or more previous backup images; and
the temporal index maintains mapping information between the saved context dataset and the one or more previous saved context datasets, and the backup image and the one or more previous backup images.