Coalescing Heterogeneous Storage Nodes via Matrix Mathematics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Heterogeneous multi-node storage systems face limitations in scaling storage capacity due to non-uniform storage node capacities, leading to gaps in matrix representation and inefficiencies in data protection and recovery.
Innovation Solution
The solution involves representing storage capacity as matrices of contiguously indexed cells and nodes, allowing for the coalescence of non-contiguous storage space and relocation of protection group members when new nodes with different capacities are added, using matrix mathematics to manage gaps and maintain continuous storage representation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If heterogeneous storage nodes with non-uniform storage capacity are added to expand storage capacity, then storage capacity increases, but gaps are created in matrix representation causing non-contiguous storage space
Solution Approach 1:
The patent merges non-contiguous storage space by coalescing cells from multiple matrices into a single contiguous matrix structure. When heterogeneous storage nodes are added, their cells are integrated with existing matrices through coalescing operations, eliminating gaps and maintaining continuous storage representation while preserving the ability to represent heterogeneous capacities.
Solution Approach 2:
The patent introduces a new dimension to the matrix representation by creating multiple matrices with different dimensions to represent storage nodes of varying capacities. These matrices are then coalesced into a unified structure that accommodates heterogeneous capacities without creating gaps in the overall storage representation.
2Stability of the object's composition
If storage nodes are indexed contiguously to maintain continuous representation, then matrix representation remains contiguous, but storage nodes with different capacities create gaps in cell allocation
Solution Approach 1:
The patent segments storage capacity into discrete cells of uniform size, allowing storage nodes with different total capacities to be represented as matrices with different numbers of cells. This segmentation enables contiguous indexing while accommodating heterogeneous capacities through variable matrix dimensions.
Solution Approach 2:
The patent changes the parameter of matrix dimensionality to accommodate varying storage node capacities. By allowing matrices to have different dimensions based on node capacity, the system maintains contiguous indexing and representation while adapting to heterogeneous storage configurations.
3Reliability
If protection group members are distributed across storage nodes for failure recovery, then reliability improves, but data relocation overhead increases when nodes are added or failed
Solution Approach 1:
The patent performs preliminary actions by pre-calculating and pre-allocating cell mappings and protection group distributions before node additions or failures occur. The coalescing operations and matrix mathematics are prepared in advance, enabling rapid data relocation and minimal disruption when nodes are added or fail.
Solution Approach 2:
The patent creates virtual copies of storage nodes and their data through the matrix representation system. Protection group members are distributed across multiple virtual representations, enabling failure recovery by copying data from surviving nodes to replace failed nodes without requiring physical data migration during the failure event.
Data Source
AI summary
A storage system has a plurality of heterogenous storage nodes characterized by non-uniform total non-volatile storage capacity. Storage capacity of all nodes is configured as same-size cells and represented as a set of matrices. The matrices have dimensions corresponding to consecutive cell indices and consecutive storage node indices. Initially, storage nodes having the same storage capacity are consecutively indexed so that the representative matrices are not ragged due to gaps, referred to herein as non-contiguous storage space, where cells do not exist because of differences in numbers of cells in adjacently indexed storage nodes. Addition of more heterogeneous storage nodes can create such gaps when the cells of those storage nodes are added to the matrices. The gaps are managed in the ragged matrix representations by coalescing the non-contiguous storage space, thereby enabling selected protection group members to be relocated using matrix mathematics such as linear algebra so that new protection groups can be created, and capacity can be scaled.