Database Cloning via Cluster Live Mounts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional database cloning approaches are slow and inefficient due to the need to move large files between source and target hosts, resulting in lengthy cloning times and significant network overhead.
Innovation Solution
A cluster-based database system implements instant recovery by live mounting directories to multiple nodes, allowing data files to be managed and recovered without physical migration, using a remote connector agent and script templates to execute recovery scripts on the host.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If large files are moved between source and target hosts for database cloning, then complete database replication is achieved, but cloning time becomes lengthy and network overhead increases
Solution Approach 1:
The patent segments the database cloning process into two distinct phases: (1) a fast metadata copy phase that replicates directory structures and file headers immediately, and (2) a background data transfer phase that moves actual file contents asynchronously. This segmentation allows the cloning operation to appear complete quickly while finishing data transfer in the background, resolving the contradiction between replication completeness and cloning speed.
Solution Approach 2:
The patent performs preliminary actions by copying metadata, directory structures, and file headers before the actual large file data transfer. This preliminary setup creates a functional clone structure that can be used immediately, while the bulk data transfer occurs in the background, thus achieving fast initial cloning without sacrificing complete data replication.
2Reliability
If large files are moved between source and target hosts for database cloning, then complete database replication is achieved, but network overhead becomes significant
Solution Approach 1:
The patent segments network traffic into two types: (1) small metadata transfer packets that establish the clone structure immediately, and (2) large data blocks transferred in the background. This segmentation reduces initial network overhead while maintaining complete replication, as the bulk data transfer occurs asynchronously and can be optimized with compression and incremental transfers.
Solution Approach 2:
The patent uses intelligent copying strategies that identify and replicate only the necessary data blocks rather than copying entire files blindly. By using checksums and block-level comparison, the system copies only changed or missing blocks, significantly reducing network overhead while ensuring complete and accurate database replication.
3Reliability
If data files are physically migrated for database recovery, then database restoration is achieved, but downtime increases
Solution Approach 1:
The patent performs preliminary database cloning and validation before failure occurs, creating a ready-to-use clone structure. When recovery is needed, the system only needs to activate the pre-prepared clone and synchronize any recent changes, rather than performing complete file migration during recovery. This preliminary preparation dramatically reduces downtime while ensuring restoration capability.
Solution Approach 2:
The patent implements dynamic recovery where the system can switch between primary and clone databases in real-time based on operational needs. The clone structure is dynamically updated with incremental changes, allowing the system to maintain service continuity by switching to the clone during recovery operations, thus minimizing downtime while ensuring complete restoration.
Data Source
AI summary
A database can be instantly cloned from a source device to a target device by a cluster mapped to a database to be cloned. Nodes of the cluster are mapped over channels to directories of the database. Scripts are generated from one or more templates that specify the order and values to be executed to perform a database job, such as cloning the database to the target device using the mappings. To clone the database, a template can be executed that generates and populates scripts, which can be executed on the target device to provide a functioning cloned database using the mapped cluster.