Storage Data Reduction by Service Type Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data reduction methods have low performance when applied to data stored using different service types, as they use a general reduction method for all types of data.
Innovation Solution
A data reduction method that selects a reduction manner matching the specific service type of the data stored in a logical unit, which includes data compression and/or deduplication, thereby improving data reduction performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a general data reduction method is used for all types of data, then the method is simple to implement, but data reduction performance is low
Solution Approach 1:
The patent segments the storage space into multiple logical units, where each logical unit is assigned a specific service type (e.g., database, file storage, log storage). Different data reduction methods are then applied to different logical units based on their service types, allowing optimized reduction performance for each data type while maintaining manageable system complexity through modular organization.
Solution Approach 2:
The patent applies the principle of local quality by tailoring the data reduction method to match the specific characteristics of each service type. For example, database data may use columnar compression while log data uses deduplication, ensuring that each local region (logical unit) has the most appropriate reduction method for its specific data characteristics, thereby maximizing overall reduction performance.
2Productivity
If different data reduction manners are applied to different service types, then data reduction performance is improved, but the complexity of determining service types increases
Solution Approach 1:
The patent performs preliminary action by pre-assigning service types to logical units before data is stored. When a logical unit is created or formatted, its service type is determined in advance based on the intended use case. This preliminary classification eliminates the need for complex real-time analysis of data characteristics during the reduction process, simplifying the detection and measurement overhead.
Data Source
AI summary
A data reduction method and apparatus are provided to improve data reduction performance. In the method a storage device obtains a first service type of first data stored in a first logical unit. The storage device processes the first data based on a first data reduction manner that matches the first service type. The storage device obtains a second service type of second data stored in a second logical unit. The second service type is different from the first service type. The storage device processes the second data based on a second data reduction manner that matches the second service type. The first data reduction manner includes data compression and/or data deduplication. The second data reduction manner includes data compression and/or data deduplication. The first data reduction manner is different from the second data reduction manner.