Distributed Network Backup Selection Mechanism
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for backing up user data in networked computer systems often overload central servers and do not efficiently distribute the load across multiple computers, especially during system upgrades.
Innovation Solution
A method where a first computer in a network sends a request to other computers in the same sub-network to identify and select the best available computer based on specific criteria such as storage space, network connections, and backup sessions, to store and retrieve user data, thereby distributing the backup process without relying on a central server.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If user data is backed up to a central server, then data storage is simplified and centralized, but the load on the server increases and network performance deteriorates
Solution Approach 1:
The patent segments the centralized backup function into distributed backup nodes across multiple computers in the network. Instead of all data flowing to a single central server, the system divides the backup workload among multiple peer computers, reducing the load on any single node and improving overall network performance during backup operations
Solution Approach 2:
The patent transitions from a single-dimension centralized backup architecture to a multi-dimensional distributed peer-to-peer backup architecture. This dimensional change allows backup operations to occur across multiple nodes simultaneously, distributing the network load and improving performance while maintaining the simplicity of automated backup management
2Productivity
If backups are distributed across multiple computers, then network load is reduced and performance improves, but system complexity increases
Solution Approach 1:
The patent implements universal backup client functionality on each peer computer in the network. Each computer is equipped with the same backup agent software that can both request backups and serve as a backup destination, eliminating the need for specialized server software and simplifying the overall system architecture while enabling distributed backup operations
Solution Approach 2:
The system employs self-service mechanisms where each peer computer automatically advertises its backup availability and capabilities to the network. Backup requests are automatically routed to appropriate peers based on their current state, eliminating the need for centralized backup management and reducing system complexity despite the distributed architecture
3Device complexity
If a central server is used for backups, then backup management is simplified, but the server becomes a single point of failure
Solution Approach 1:
The patent segments the backup storage function across multiple independent peer computers in the network, eliminating the single point of failure inherent in centralized server backups. Each peer acts as an independent backup node, so if one fails, backups can still be stored and retrieved from other peers, significantly improving backup availability and reliability
4Reliability
If multiple computers are used for backups, then load is distributed and reliability improves, but selecting the best computer increases complexity
Solution Approach 1:
The patent implements a feedback mechanism where each peer computer continuously advertises its backup availability status, storage capacity, and current load to the network. The backup requesting computer receives this feedback information and automatically selects the most appropriate peer based on real-time network conditions, simplifying the selection process while ensuring optimal backup placement and maintaining high reliability
Data Source
AI summary
User data stored on a first computer in a network is backed up to a computer which is one of a plurality of other computers in the network. The user data may be backed-up because a new operating system is to be installed. A request message indicating that the user data needs to be backed up is transmitted from the first computer to the network and is received at the other computers. One of the other computers is selected based on data indicating the extent to which each of the replying computers complies with one or more selection criteria. The first computer then transmits the user date to the selected computer with an identifier identifying the user data. The user data is stored in storage associated with the selected computer. The first and other computers may be in the same sub-network. The selection may be made by the first computer.