Shared Memory Network Links for High Availability Data Transfer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data transfer solutions in shared storage environments face bandwidth limitations and stability issues, leading to performance bottlenecks and a lack of high availability due to inadequate failover mechanisms.
Innovation Solution
A method and system for configuring and managing network links between hardware nodes, including software agents that dynamically assess and prioritize network connections, segment data files for efficient transfer, and implement automatic failover to ensure continuous data transfer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If standard shared storage and distributed file systems are used for data transfer, then data storage capability is provided, but data transfer speed is limited to standard speed with lag times
Solution Approach 1:
The patent segments data files into multiple data segments and transfers them simultaneously through multiple network links between hardware nodes. This parallel transmission approach eliminates the sequential transfer limitation of standard file systems, significantly reducing transfer time and increasing effective data transfer speed across the network storage system.
Solution Approach 2:
The patent transitions from single-channel sequential data transfer to multi-dimensional parallel data transfer by establishing multiple network links between hardware nodes. This dimensional expansion allows simultaneous data flow through different paths, breaking the standard speed limitation and reducing transfer lag times.
2Reliability
If single network links are used for data transfer, then system complexity is reduced, but reliability is compromised due to lack of failover mechanisms
Solution Approach 1:
The patent pre-configures multiple network links and establishes connectivity between hardware nodes before data transfer operations begin. This preliminary setup includes creating data segment files and preparing alternative transfer paths, so that when a link fails, the system can immediately switch to a backup link without interruption, ensuring high availability while managing complexity through advance preparation.
Solution Approach 2:
The patent implements redundancy by maintaining multiple active network links and backup transfer paths before failures occur. This cushioning approach ensures that if one network link malfunctions, alternative paths are already in place to maintain data transfer continuity, thereby improving reliability without requiring complex real-time decision-making during failures.
3Productivity
If data is transferred through multiple network links in parallel, then data transfer speed is improved, but bandwidth management complexity increases
Solution Approach 1:
The patent implements self-service bandwidth management where each hardware node autonomously monitors its own network link utilization and automatically adjusts data segment transfer rates. The system self-regulates bandwidth allocation based on current network conditions without requiring external management intervention, thereby maintaining high throughput while simplifying bandwidth management complexity.
Solution Approach 2:
The patent incorporates feedback mechanisms where hardware nodes continuously monitor network link performance and transfer rates. Based on this feedback, the system dynamically adjusts the allocation of data segments to different network links, optimizing bandwidth utilization and maintaining high productivity while automatically managing the complexity of multi-link coordination.
Data Source
AI summary
A method and system for improving a shared memory file transfer setup and usage is provided. The method includes configuring network links connected between hardware nodes. For every shared storage link, each hardware node includes a command section and a data storage section. A storage capacity for a hardware disk drive is determined based on a number of the hardware nodes and the hardware disk drive is connected to the hardware nodes. A software agent is installed within each of the hardware nodes and properties for the network links are configured. Each software agent is enabled and a connectivity status, bandwidth, and loading condition are tested for each of the network links. Each hardware node is registered with the hardware disk drive and communications between the hardware disk drive and the hardware nodes are enabled via the network links.