Storage Auto-Scale Processing for Dynamic Cloud Configuration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in dynamically changing the storage apparatus configuration to scale with cloud server load changes, leading to inefficient resource utilization and adverse impacts on business activities due to wasteful CPU and memory consumption during replication.
Innovation Solution
An information processing apparatus and method that selectively replicates high-frequency access data and specific host localization data, dynamically adjusting storage resources such as ports, processors, and cache memory by using a storage auto-scale processing apparatus to manage access frequency and distribute loads across different RAID groups.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If all data within the logical volume is replicated to enhance storage performance in response to host scale-out, then storage performance is improved, but CPU and memory resources are wastefully consumed and replication time increases
Solution Approach 1:
The patent applies local quality by differentiating data into hot data (high access frequency) and cold data (low access frequency), and applying replication only to hot data. This selective approach ensures that resources are concentrated on replicating only the most accessed data portions, thereby improving storage performance while avoiding wasteful consumption of CPU and memory resources on cold data that does not require frequent replication.
Solution Approach 2:
The patent implements partial action by replicating only a subset of data (hot data with access frequency above a threshold) rather than all data in the logical volume. This partial replication strategy achieves sufficient storage performance enhancement for actively accessed data while significantly reducing the computational overhead and resource consumption associated with replicating entire volumes including cold data.
2Productivity
If all data within the logical volume is replicated to enhance storage performance in response to host scale-out, then storage performance is improved, but the time required for replication execution increases
Solution Approach 1:
The patent applies local quality by differentiating data into hot data (high access frequency) and cold data (low access frequency), and applying replication only to hot data. This selective approach ensures that resources are concentrated on replicating only the most accessed data portions, thereby improving storage performance while avoiding wasteful consumption of CPU and memory resources on cold data that does not require frequent replication.
Solution Approach 2:
The patent implements partial action by replicating only a subset of data (hot data with access frequency above a threshold) rather than all data in the logical volume. This partial replication strategy achieves sufficient storage performance enhancement for actively accessed data while significantly reducing the computational overhead and resource consumption associated with replicating entire volumes including cold data.
3Device complexity
If the storage apparatus does not have a redundant configuration, then device complexity is reduced, but the ability to dynamically change configuration in response to scale-out is limited
Solution Approach 1:
The patent applies dynamics by implementing a dynamic data replication mechanism that automatically activates when host scale-out is detected. The system continuously monitors access frequencies and dynamically adjusts replication based on current data characteristics and workload demands, allowing the storage apparatus to adapt its configuration in response to scale-out events without requiring permanent redundant hardware components.
Solution Approach 2:
The patent uses copying by creating replica copies of hot data from the primary logical volume to secondary storage locations. This copying mechanism enables dynamic configuration changes and data protection without requiring permanent redundant hardware, as replicas are created on-demand based on access frequency thresholds and can be dynamically adjusted or removed when no longer needed.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An information processing apparatus and method capable of minimizing influences affecting business activities and dynamically changing the configuration of a storage apparatus in response to scale-out of hosts are proposed. The information processing apparatus for managing the configuration of the storage apparatus which constitutes a hybrid cloud system is provided with a data acquisition unit and a data copy management unit; and if the data acquisition unit detects the scale-out of the hosts and determines that a first volume, which stores data from the hosts, in the storage apparatus has a high load, the data acquisition unit issues an instruction to the data copy management unit to copy the data and the data copy management unit issues an instruction to the storage apparatus, in accordance with the instruction from the data acquisition unit, to copy the data, regarding which the access frequency from each host is high, to a second volume and to copy the data, regarding which the access frequency from a specified host is high, to a third volume.