Searchable Index for Backup Data Version Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data retrieval systems face difficulties in efficiently locating specific versions of data objects across multiple backup sources, requiring extensive searching and restoration of data sets, which is time-consuming and inefficient.
Innovation Solution
A method is introduced that processes backup and production data to generate a searchable index, associating related data objects based on common file names, paths, or identifiers, allowing for efficient searching and retrieval of data object versions by using this index to determine responsive data objects and display their version relationships.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple individual backup sources are searched and restored to locate a desired version, then the desired version can be found, but the search process becomes time-consuming and inefficient
Solution Approach 1:
The patent applies preliminary action by creating and maintaining an index of backup data in advance. The index contains metadata about all backup sources, including file names, paths, timestamps, and storage locations. This pre-processing allows the search function to quickly locate desired versions without actually restoring and searching through multiple backup sources at the time of the search operation.
Solution Approach 2:
The patent introduces an index as an intermediary data structure between the backup sources and the search function. This index acts as a mediator that stores summarized information about backup data, allowing the search to be performed on the compact index rather than on the actual backup data itself. The index contains pointers to the actual backup sources, enabling efficient navigation to the desired version.
2Difficulty of detecting and measuring
If backup data is restored to search for desired files, then files can be located, but the process requires extensive restoration and browsing operations
Solution Approach 1:
The patent extracts the essential search-relevant information from the backup data and stores it separately in an index. Instead of restoring entire backup datasets to perform searches, the system extracts and indexes metadata such as file names, paths, timestamps, and storage locations. This extraction allows searches to be performed on the extracted index data without involving the complex restoration process.
Solution Approach 2:
The patent creates a copy of the backup data's metadata and organizational structure in the form of an index. This index is a simplified representation that mirrors the backup data's hierarchy and content information but occupies minimal space. The search operation works on this copied index structure rather than the original backup data, dramatically simplifying the search process.
3Loss of information
If multiple backup sources are individually searched, then all versions can be found, but the relationship between files must be determined separately
Solution Approach 1:
The patent merges information from multiple backup sources into a unified index structure. The index consolidates metadata about files across all backup sources, maintaining relationships between different versions of the same file. By combining this unified index with the search function, the system can determine file relationships automatically during the search process, eliminating the need for separate relationship analysis operations.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
Searching data is disclosed. A searchable index is used to determine which, if any, data objects comprising a set of backup data are responsive to a search query. For the search query, a search result is generated. The search result includes, in the event a first data object that is responsive to the search query comprises a version of one or more other data objects that are responsive to the search query, data indicating that the first data object comprises a version of said one or more other data objects.