Storage Node Logical Capacity Setting via I/O Performance Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In distributed data storage systems, incorporating storage nodes with different capacities and I/O performances poses challenges in evenly distributing data storage capacity and data access I/O numbers, leading to potential performance bottlenecks and inefficient resource utilization.
Innovation Solution
A method is introduced to set logical capacities of storage regions based on I/O performance values, where a reference storage node is designated to divide other nodes into first and second storage regions, matching their I/O performance, and system configuration information is updated to reflect these changes, ensuring balanced resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If storage nodes with different capacities and I/O performances are incorporated into the distributed data storage system, then the system can accommodate diverse storage devices and expand flexibility, but the data storage capacity and data access I/O numbers cannot be evenly distributed, leading to performance bottlenecks
Solution Approach 1:
The patent applies local quality by setting different logical capacities for storage regions based on the specific I/O performance characteristics of each storage node. Instead of uniform allocation, each node's storage regions are configured with capacities tailored to its I/O capabilities, ensuring that nodes with higher I/O performance handle larger data access loads while maintaining even distribution across the system
Solution Approach 2:
The patent changes the parameter of logical capacity allocation from uniform to variable based on I/O performance measurements. By dynamically adjusting the logical capacity parameters of storage regions according to each node's I/O performance characteristics, the system achieves even distribution of data storage capacity and data access I/O numbers despite incorporating storage nodes with different performance levels
2Productivity
If logical capacities of storage regions are set based on I/O performance values, then even distribution of data storage capacity and data access I/O numbers is achieved, but the system complexity increases due to additional configuration management
Solution Approach 1:
The patent implements self-service by enabling storage nodes to automatically measure their own I/O performance and use these measurements to determine their logical capacity allocations. This automated self-configuration approach reduces manual intervention and simplifies management complexity while achieving even distribution of data storage capacity and data access I/O numbers based on actual performance characteristics
3Device complexity
If storage nodes are constituted using a single server, then the system structure is simple, but the storage capacity cannot be expanded and the node becomes a single failure point
Solution Approach 1:
The patent applies segmentation by dividing the storage system into multiple independent storage nodes that can be distributed across different servers. This segmentation eliminates the single failure point vulnerability while maintaining relatively simple structures at each node level. Each storage node operates independently, so failures are isolated and do not affect the entire system
4Reliability
If distributed data storage system duplicates data across multiple nodes, then system availability is maintained during node failures, but the storage capacity requirement increases
Solution Approach 1:
The patent changes the parameter of logical capacity allocation to account for duplication requirements while optimizing total storage capacity usage. By setting logical capacities based on I/O performance and distribution requirements, the system maintains necessary data copies for availability while reducing redundant storage compared to uniform duplication strategies, thereby optimizing the quantity of storage capacity required
Data Source
AI summary
A storage node administration device enables setting a logical capacity of a storage region based on an I/O performance value of a storage node. In a distributed data storage system, the storage node administration device includes a setting unit setting at least one reference node from a plurality of storage nodes having different I/O performances, dividing a storage region of the storage nodes other than the reference node into a first storage region and a second storage region, referring to the I/O performance and the logical capacity of the reference node, and setting the logical capacity of each region to match the I/O performance of the storage nodes other than the reference node; a determination unit determining system configuration information which reflects a configuration change of the distributed data storage system; and a transmission unit transmitting the system configuration information to the storage nodes having different I/O performances.