Multi-Node Storage Volume Creation via Local Pool Capacity Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multi-node storage systems, the capacity provided by individual nodes is limited, leading to write errors when creating volumes larger than the capacity of a single node, as the global pool capacity exceeds the combined capacity of the nodes, causing inefficiencies and errors in data storage.
Innovation Solution
A storage system is implemented where a global pool is managed by multiple local pools across storage nodes, with a storage management unit ensuring that the used capacity of created volumes remains equal to or less than the available capacity of the local pool, preventing write errors by dynamically managing capacity and selecting appropriate nodes for volume creation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If a volume capacity larger than a single node's capacity is created in a multi-node storage system, then the global pool capacity is utilized more effectively, but write errors occur when the data amount exceeds the creating node's capacity
Solution Approach 1:
The patent segments the storage system into multiple nodes, each with its own local pool. The global pool is formed by aggregating these segmented local pools. When creating a volume, the system selects a specific node (segment) to create the volume, ensuring the volume capacity does not exceed that node's local pool capacity, thereby preventing write errors while still utilizing the global pool's total capacity across multiple segments.
2Adaptability or versatility
If the storage system presents a large global pool capacity to users, then storage versatility is improved, but the limited capacity of individual nodes causes write errors
Solution Approach 1:
The patent introduces a storage management unit as an intermediary between the user and the physical storage nodes. This intermediary presents the aggregated global pool capacity to users, allowing flexible volume creation up to the total available capacity. The management unit then mediates the volume creation process by selecting an appropriate node and ensuring the volume capacity fits within that node's local pool, thus maintaining both user-facing versatility and backend reliability.
3Ease of operation
If volumes are created without considering individual node capacity constraints, then storage system simplicity is maintained, but write errors occur due to capacity mismatches
Solution Approach 1:
The patent implements preliminary action by having the storage management unit pre-select an appropriate node for volume creation based on the requested volume capacity and the available capacity of each node's local pool. This preliminary selection and validation occur before the volume creation is executed, ensuring that the chosen node can accommodate the volume without exceeding its local pool capacity. This approach maintains operational simplicity for users while preventing write errors through advance capacity verification.
Data Source
AI summary
In a multi-node storage system, a node's capacity has an upper limit, and capacities provided by nodes are smaller than a capacity of a global pool. A volume having a capacity larger than the capacity of one node is created by the node. A write error occurs when an amount of data larger than the capacity of the node is written. A storage system reduces the frequency of such a write error. A global pool is based on a plurality of local pools of a plurality of storage nodes that constitute a node group. In any of the storage nodes, a capacity relationship is maintained where a used capacity of a volume created by the storage node is equal to or less than an available capacity of a local pool of the storage node. A storage management unit manages the node group and selects the storage node.