Data Control Device Caching for Distributed Storage Performance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for data utilization between sites, such as hybrid cloud and edge-core cooperation, face performance issues due to the need for data transmission, particularly in applications where file virtualization is used, as seen in techniques like US-A-2009/0125634, which are limited to media reproduction applications with fixed bit rates.
Innovation Solution
A data control device that manages data units across sites by specifying access circumstances and controlling caching before application execution, using a processor to optimize data access and reduce performance degradation in applications that use data units from other sites.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If data is transmitted between sites for file virtualization, then data sharing capability is improved, but application performance deteriorates
Solution Approach 1:
The patent applies preliminary action by caching data from other sites at the own site before the application actually needs it. The storage device proactively retrieves and stores data units from remote storage devices, so that when the application accesses the data, it can be served from the local cache rather than transmitted over the network, thus improving performance while maintaining data sharing capability
Solution Approach 2:
The patent introduces a caching mechanism as an intermediary between the application and the remote storage devices. This cache acts as a buffer that stores copies of data units, allowing the application to access data locally without direct communication with remote sites, thereby reducing network transmission overhead while preserving the ability to share data across sites
2Productivity
If caching is performed for data units from other sites, then application performance is improved, but network bandwidth consumption increases
Solution Approach 1:
The patent dynamically adjusts caching parameters such as cache size, caching duration, and data selection based on access patterns and application behavior. By changing these parameters adaptively, the system optimizes the balance between performance improvement and network bandwidth consumption, caching only the most frequently accessed data units for appropriate durations
Solution Approach 2:
The patent applies partial action by caching only specific portions of data rather than entire files, and by caching data selectively based on predicted access patterns. This approach improves performance for the most critical data access scenarios while minimizing unnecessary network bandwidth consumption for data that wouldn't be accessed anyway
Data Source
AI summary
In a management node that manages a distributed file and object storage that accessibly manages a file used by an application, the distributed file and object storage is accessible to a file managed by a storage of another site, and a management node includes a processor, and the processor is configured to specify an access circumstance relating to a file by an application, and control caching by the distributed file and object storage of the own site with respect to the file managed by the storage of the other site used by the application, before the application is executed, based on the access circumstance.