Data Restoration Using Parent Node Identifier Merging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data restoration methods are inefficient and resource-intensive, particularly when dealing with multiple nodes, as they generate numerous streams for each node, leading to increased time consumption and resource waste in the backend backup server.
Innovation Solution
The method involves obtaining an index related to data blocks and determining a node identifier set, where child node identifiers are replaced with a father node identifier, reducing the number of streams sent to the backup server and optimizing metadata retrieval and resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If child node identifiers are used for each node in the index, then data blocks can be precisely retrieved, but the number of streams increases and resource consumption increases
Solution Approach 1:
Multiple child node identifiers are merged into a single parent node identifier in the restoration request. The backup server receives one stream containing data blocks for multiple child nodes, consolidating what would otherwise require multiple separate streams. This reduces the number of streams from N (number of child nodes) to 1, improving restoration efficiency while maintaining the ability to retrieve precise data blocks through the parent node's data block range information.
2Measurement precision
If child node identifiers are used for each node in the index, then specific data blocks can be obtained, but time consumption for target validation increases
Solution Approach 1:
The validation process is optimized by validating the parent node identifier once instead of validating multiple child node identifiers separately. The backup server validates the parent node's existence and retrieves its data block range information, then uses this information to identify the specific data blocks for all child nodes. This reduces validation time from O(N) to O(1) where N is the number of child nodes.
3Measurement precision
If child node identifiers are sent to the backup server, then data blocks can be retrieved accurately, but resource consumption in the backup server increases
Solution Approach 1:
The restoration request consolidates multiple child node identifiers into a single parent node identifier, reducing the number of lookup operations required in the backup server. The backup server performs one lookup for the parent node to obtain the data block range, then identifies all relevant data blocks within that range. This approach reduces computational overhead and resource consumption compared to performing N separate lookups for N child nodes, while maintaining accurate data block retrieval.
Data Source
AI summary
Embodiments of the present disclosure relates to restoring data. The embodiments comprise obtaining an index related to a group of data blocks, determining a node identifier set of a plurality of nodes in the index corresponding to a plurality of data blocks to be restored, determining, based on the index and the node identifier set, whether a first group of child node identifiers in the node identifier set related to a first group of child nodes of the plurality of nodes is to be replaced with a father node identifier of a father node of the first group of child nodes, and in response to replacing the first group of child node identifiers with the father node identifier, sending, to the backup server, a request for obtaining the plurality of data blocks to obtain the plurality of data blocks from the backup server, the request comprising the father node identifier.