VM Backup File Indexing via Live Browse
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing solutions for browsing backed-up virtual machine data lack efficient indexing capabilities, requiring users to specify the source VM and backup copy, and do not persist information about backed-up files, making it difficult to search and access data without knowledge of the original VM environment and backup storage.
Innovation Solution
A streamlined file indexing approach that creates a hypervisor-free live browse feature for VM block-level backup copies, allowing for indexing of file names, directory structures, and file contents, enabling searchable indexes without mounting the backup copies, and storing these indexes in a persistent data structure for easy retrieval.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If block-level backup copies are created without file-level indexing, then backup operations remain efficient and hypervisor-independent, but users cannot search or access backed-up files without knowing the source VM and backup copy identity
Solution Approach 1:
The system performs indexing operations during the backup process itself, creating file name indexes and content indexes before the backup data is stored. This preliminary action allows users to search and access files later without needing to know the source VM or backup copy identity, while the indexing overhead is absorbed during the backup window rather than impacting production operations.
Solution Approach 2:
The patent introduces indexing structures (file name indexes and content indexes) as intermediary layers between the raw backup data and user search queries. These indexes act as mediators that translate user-friendly search terms into locateable backup data, eliminating the need for users to specify source VM details while maintaining efficient access to backed-up files.
2Productivity
If users must specify source VM and backup copy identity to access data, then the system maintains simple storage structure, but user productivity decreases due to time-consuming data location and search operations
Solution Approach 1:
The system creates comprehensive indexes of file names and file contents during backup operations, so that when users need to search for data, the information is already organized and ready for instant retrieval. This eliminates the time-consuming process of manually browsing through backup copies and locating specific files.
Solution Approach 2:
The indexing system provides immediate feedback to users about the contents of backup data through searchable indexes. Users can query for specific files or content, and the system rapidly returns relevant results with location information, enabling efficient data retrieval without requiring users to specify source VM details.
3Adaptability or versatility
If file indexing is performed on all backup data, then search capability is enhanced, but storage space consumption increases due to redundant index data
Solution Approach 1:
The patent segments the indexing process into two distinct indexes: a file name index that maps filenames to backup locations, and a content index that stores searchable content keywords. This segmentation allows the system to provide comprehensive search capability while storing only the necessary indexing information rather than duplicating the entire backup data, thereby reducing storage space consumption.
Solution Approach 2:
The system extracts only the essential indexing information (file names, paths, and content keywords) from the backup data and stores this metadata separately in the indexing structures. By taking out only the necessary indexing elements rather than storing complete data copies, the system achieves enhanced search capability while minimizing additional storage space requirements.
Data Source
AI summary
An illustrative file indexing approach enhances what was previously possible with hypervisor-free live browsing of virtual machine (VM) block-level backup copies. Capabilities are described for indexing files discovered in VM block-level backup copies, including indexing of directory structures and file content. The illustrative file indexing functionality activates a live-browse session to discover files present within VM block-level backup copies and indexes file names and directory structures as created by an original source VM, resulting in an illustrative file index. The illustrative file indexing functionality optionally indexes file contents within VM block-level backup copies, resulting in an illustrative content index. The file index and content index are retained in persistent data structure(s) stored apart from the VM block-level backup copies. The indexes are searchable without mounting or live-browsing the VM block-level backup copies. In some embodiments the file index and the content index are consolidated.