Controller-Specific RAID Address Spaces for Flexible IO Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing RAID systems face performance degradation due to dependency on logical volume address mapping, leading to frequent stripe changes and reduced flexibility and efficiency in IO processing.
Innovation Solution
The method involves partitioning RAID layer address spaces independently, using a data cache to synchronize IO data between controllers, and performing write operations based on pre-partitioned RAID layer spaces, decoupling logical volume and RAID layer address affiliations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If address space partitioning is performed at the logical volume layer, then resource access conflicts between controllers are avoided, but the RAID layer becomes dependent on logical volume address mapping and cannot operate independently
Solution Approach 1:
The patent segments the address space management into two independent layers: logical volume layer address space and RAID layer address space. Each layer maintains its own address mapping relationships independently, allowing the RAID layer to operate autonomously without depending on logical volume address mapping changes.
Solution Approach 2:
The patent introduces an address translation mechanism that acts as an intermediary between the logical volume layer and RAID layer. This intermediary layer enables independent address space management at each level while maintaining proper mapping relationships, resolving the dependency issue.
2Reliability
If the RAID layer follows address partitioning rules provided by the logical volume, then address space isolation is achieved, but frequent stripe changes occur when spatial affiliation partitioning changes, reducing IO processing flexibility and efficiency
Solution Approach 1:
The patent divides address space management into separate logical volume layer and RAID layer, each with independent address mapping. This segmentation allows the RAID layer to maintain stable address mappings even when logical volume partitioning changes, eliminating frequent stripe changes and improving IO processing efficiency.
Solution Approach 2:
The patent pre-establishes independent address mapping relationships at the RAID layer that are decoupled from logical volume address changes. This preliminary setup ensures that when logical volume partitioning changes, the RAID layer does not need to perform frequent address re-mapping operations.
3Productivity
If multiple controllers access the same address space, then performance overlay is achieved, but resource access conflicts arise
Solution Approach 1:
The patent segments the address space into controller-specific portions at the RAID layer, where each controller has dedicated address ranges. This segmentation enables multiple controllers to operate in parallel without accessing the same physical resources simultaneously, eliminating resource access conflicts while maintaining performance overlay.
Data Source
AI summary
An IO processing method and apparatus, a device, and a storage medium are disclosed. The method includes: acquiring IO data sent by a user by means of a logical volume, and sending, by using the logical volume, the IO data to a data cache by a first controller; allocating a second controller to the IO data by a RAID, and synchronizing the IO data from the first controller to the second controller by the data cache; and determining a target address space according to a pre-partitioned RAID layer address space by the second controller if the IO data is a data write request, and performing a data write operation on the target address space based on the IO data, where the pre-partitioned RAID layer address space is a RAID layer address space generated after partitioning a corresponding address space for each controller at a RAID layer.