Storage Redundancy via Logical Chunk Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high cost of expanding capacity in storage systems configured with RAID groups, particularly in all flash arrays and mid-range systems with a small number of drives, due to the need for maintaining redundancy and the initial high cost of implementing distributed RAID methods.
Innovation Solution
A computer system that includes d physical storage devices, where d is an integer of 2 or more, creates n×v×d physical storage areas, maps them into v×d logical chunks, and adjusts mapping information to add redundancy by writing user data and redundant data across n pieces of physical storage areas, allowing for flexible extension of physical storage devices while maintaining redundancy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If drives are extended in units of the number of drives configuring the RAID group to ensure redundancy, then reliability is maintained, but device cost increases
Solution Approach 1:
The patent segments the RAID group into multiple independent RAID groups, each with its own redundancy. This allows individual drives to be extended without requiring extension of the entire RAID group, thereby reducing the cost of capacity addition while maintaining redundancy within each segment. The mapping information is updated to reflect the new segmented structure, enabling flexible drive extension at the level of individual drives or small groups rather than requiring extension of all drives in the RAID group.
2Quantity of substance
If distributed RAID method is adopted, then storage capacity is improved, but device cost increases due to minimum two parity groups requirement
Solution Approach 1:
The patent divides the storage system into multiple independent RAID groups that can operate separately. This segmentation allows the system to start with a single RAID group containing one parity group, reducing the initial introduction cost. As storage capacity needs grow, additional RAID groups can be added incrementally. Each RAID group maintains its own mapping information and redundancy structure, enabling flexible capacity expansion without requiring a minimum of two parity groups from the outset.
3Reliability
If RAID group extension is performed to maintain redundancy, then reliability is preserved, but productivity decreases due to high expansion cost
Solution Approach 1:
The patent implements a dynamic mapping information update mechanism that adapts to different drive extension scenarios. When drives are extended, the system dynamically updates mapping information to associate new physical storage areas with logical chunks, allowing flexible capacity expansion. This dynamic approach enables the system to maintain redundancy through selective RAID group extension rather than requiring extension of all drives, thereby improving capacity expansion efficiency and productivity while preserving reliability.
Data Source
AI summary
Securing redundancy for physical storage devices that are extended in units smaller than physical storage devices configuring one RAID group. When d+r pieces of physical storage devices are connected by connecting r pieces of physical storage devices, a computer: adds v×r pieces of logical chunks; adds n×v pieces of physical storage areas in each additional storage device; changes mapping information to associate n pieces of physical storage areas with v×(d+r) pieces of logical chunks under a mapping condition; in response to a write request of user data, creates redundant data; determines a first logical chunk corresponding to the write request; and respectively writes n pieces of element data including the user data and the redundant data into n pieces of physical storage areas corresponding to the first logical chunk, based on the mapping information.