Logical Unit Creation via Data Block Referencing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In data storage systems, creating new logical units often involves costly data copying and bandwidth usage, as well as unnecessary disk space allocation, especially in distributed virtual infrastructures where storage demand fluctuates.
Innovation Solution
The logical unit management module identifies source data blocks from existing logical units and creates a new logical unit by referencing these blocks, rather than copying the data, thereby reducing CPU and bandwidth costs and preserving disk space.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is copied from existing LU to newly created LU, then data integrity is ensured, but data processing costs and bandwidth usage increase significantly
Solution Approach 1:
The patent implements a hybrid approach: it uses copying only when necessary (when source LU data is modified) rather than always copying at creation. The system creates thin-provisioned LUs that reference source data without immediate copying, and only copies data when the source LU undergoes modifications, thus reducing unnecessary data processing costs while maintaining data integrity when needed.
Solution Approach 2:
The system performs preliminary setup by creating logical unit mappings and establishing references to source data blocks before any actual data copying occurs. This preliminary configuration allows the system to defer data copying until it is actually needed, reducing initial data processing overhead while ensuring data integrity is maintained through pre-established reference relationships.
2Reliability
If data is copied from existing LU to newly created LU, then data independence is achieved, but transfer bandwidth costs increase
Solution Approach 1:
The patent implements selective copying based on data modification events. When a source LU is created, the new LU establishes a reference to the source data without immediate copying. Data is copied only when the source LU data is modified, ensuring data independence is achieved at the appropriate moment while minimizing unnecessary bandwidth consumption during LU creation operations.
3Reliability
If data is copied from existing LU to newly created LU, then data isolation is ensured, but disk space consumption increases
Solution Approach 1:
The patent merges multiple logical units into a single physical storage space through thin provisioning. Multiple LUs share the same physical disk blocks, allowing data isolation to be maintained at the logical level while physical storage is consolidated. This approach significantly reduces disk space consumption compared to traditional one-to-one mapping where each LU requires dedicated physical storage.
Solution Approach 2:
The system uses copying only when data isolation is actually required (upon source LU modification) rather than maintaining permanent copies. This event-driven copying approach ensures data isolation is achieved when necessary while minimizing disk space consumption by avoiding redundant copies during normal LU operations.
4Productivity
If thin provisioning is implemented without data copying, then resource efficiency improves, but data consistency may be compromised
Solution Approach 1:
The patent implements a feedback mechanism that monitors source LU modifications. When the source LU data changes, the system detects this modification and triggers appropriate actions (such as copying data or updating references) to maintain data consistency. This feedback-driven approach ensures resource efficiency through thin provisioning while preventing data consistency issues by responding to actual modification events.
Data Source
AI summary
Techniques for creating logical units associated with a data storage system. In one example, a method comprises the following steps. One or more blocks of source data associated with a data storage system are identified. The one or more blocks of source data are associated with at least a first logical unit that exists in the data storage system. A second logical unit is created which references the one or more identified blocks of source data.