Virtual Unit Set Management for Cloud Storage Overhead Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional elastic cloud storage systems face high computing load pressure on physical nodes and incur significant overhead due to the need for data re-calculation and re-storage when virtual units are split or merged, leading to inefficient resource allocation and potential data integrity issues.
Innovation Solution
The implementation of a method that uses dynamic two-level hashing to manage virtual units, allowing for easy splitting and merging of sets of virtual units, thereby optimizing computing resource allocation and reducing overhead by sharing resources among virtual units within sets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If virtual units are split or merged to adjust storage capacity, then storage system flexibility is improved, but computing overhead increases due to data re-calculation and re-storage
Solution Approach 1:
The patent segments the storage system into multiple sets of virtual units, where each set can be independently managed. When scaling occurs, only the necessary sets are split or merged rather than requiring system-wide virtual unit reconfiguration. This segmentation allows flexible storage capacity adjustment while minimizing the scope of data re-calculation and re-storage operations.
Solution Approach 2:
The patent pre-calculates and stores hash mapping relationships between objects and virtual units before scaling operations. When virtual units need to be split or merged, the system uses these pre-computed mappings to quickly determine data redistribution requirements, avoiding the need to recalculate all hash values from scratch and significantly reducing computing overhead.
2Power
If computing resources are allocated to each virtual unit for data processing, then data processing capability is improved, but physical nodes experience heavy computing load pressure
Solution Approach 1:
The patent merges multiple virtual units into sets that share common computing resources. Instead of allocating separate computing resources to each individual virtual unit, the system consolidates resource allocation at the set level, allowing virtual units within the same set to share computing capabilities. This merging approach maintains data processing capability while significantly reducing the total computing load pressure on physical nodes.
Solution Approach 2:
The patent implements universal computing resource pools that can serve multiple virtual units within a set. These shared computing resources perform data processing tasks for any virtual unit in the set, eliminating the need for dedicated computing resources per virtual unit. This multi-functional approach optimizes resource utilization and reduces overall computing load pressure on the system.
3Manufacturing precision
If the number of virtual units is increased to improve storage granularity, then storage precision is improved, but physical nodes require enormous computing resources
Solution Approach 1:
The patent introduces a hierarchical dimension to virtual unit management by organizing virtual units into sets. This hierarchical structure allows the system to achieve fine storage granularity through the virtual unit level while managing computing resources at the higher set level. The dimensional hierarchy decouples storage precision requirements from computing resource requirements, enabling high granularity without proportional increases in computing resource consumption.
Solution Approach 2:
The patent uses shared hash mapping structures that can be copied and applied across multiple virtual units within a set. Instead of maintaining separate computing resources for each virtual unit, the system copies and reuses the same hash mapping and resource allocation patterns across virtual units in the same set, achieving high storage granularity with minimal additional computing resources.
Data Source
AI summary
Techniques for processing an access request and updating a storage system are provided. For instance, a method comprises: receiving an access request for an object associated with a storage system, the storage system including a plurality of physical nodes, each of the plurality of physical nodes including at least one set of virtual units, each set of virtual units including at least one virtual unit; determining, from a plurality of sets of virtual units included in the plurality of physical nodes of the storage system, a target set of virtual units associated with the object; and determining, from the target set of virtual units, a target virtual unit corresponding to the object. With the technical solution of the present disclosure, not only a set of virtual units on a physical node may be easily split and merged, but also huge computing resources that need to be allocated may be saved, so better user experience may be brought about at a lower cost.