Dispersed Storage Data Placement Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data sharing solutions in virtualized server environments lack scalability, performance, and cost-effectiveness, as they rely on networked storage and distributed file systems that are not optimized for high availability and disaster recovery.
Innovation Solution
A computer-implemented method that utilizes dispersed storage devices to create a global storage pool by optimizing file placement across local storage pools, allowing multiple servers to access and manage data efficiently, with file images distributed based on optimization criteria to enhance scalability and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional networked storage solutions are used for data sharing in virtualized environments, then data sharing capability is provided, but scalability and performance are limited
Solution Approach 1:
The patent segments the global storage pool into multiple distributed file images across different local storage pools. Each server maintains file images locally, eliminating the single-point bottleneck of centralized networked storage. This segmentation enables parallel access and improves scalability while maintaining data sharing capabilities.
Solution Approach 2:
The patent transitions from a centralized storage architecture to a distributed architecture by adding the dimension of spatial distribution across multiple servers. File images are replicated across different physical locations (servers), enabling concurrent access from multiple virtual machines simultaneously, thus improving performance and scalability.
2Ease of operation
If distributed file systems are used for shared storage, then data accessibility is improved, but resource consumption and inter-server bandwidth usage increase
Solution Approach 1:
The system performs preliminary actions by pre-creating file images in local storage pools before they are needed. When a virtual machine needs access to a file, the image is already available locally, eliminating the need for real-time network transfers and reducing bandwidth consumption while maintaining fast accessibility.
Solution Approach 2:
The patent creates copies of file images in local storage pools across different servers. Instead of accessing the original file over the network, virtual machines access local copies, which reduces inter-server bandwidth usage while maintaining data accessibility and availability.
3Productivity
If file images are distributed across local storage pools, then scalability is improved, but system complexity increases
Solution Approach 1:
The patent introduces a storage manager as an intermediary that automatically handles the complexity of file image distribution, replication, and retrieval. The storage manager abstracts the distributed storage complexity from users and applications, allowing the system to scale without proportionally increasing operational complexity.
Solution Approach 2:
The system implements self-service mechanisms where the storage manager automatically creates, distributes, and manages file images across local storage pools without manual intervention. This automation handles the complexity internally while presenting a simple interface to users, enabling scalability without proportional increases in operational burden.
Data Source
AI summary
A storage system provides shared storage by utilizing dispersed storage devices while optimizing both the placement of data across the dispersed storage devices and the method for accessing the stored data. The storage system enables high level of scalability and performance while minimizing resource consumption.