Distributed Backup Media Server Using Agent Relay
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current backup and restore techniques for distributed applications become impractical as they grow in size, especially across various networked computer systems and geographic locations, due to the complexity and inefficiency of managing data across multiple servers and storage media.
Innovation Solution
A system comprising a media server and multiple servers, where agents on each server provide control information to establish network connections for data backup and restore operations, allowing the media server to efficiently collect and store data from multiple sources using parallel network connections, reducing the need for extensive intelligence at the media server about all associated servers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If current backup techniques are used for distributed applications, then backup operations can be performed, but the complexity and inefficiency increase as the distributed application grows in size and geographic distribution
Solution Approach 1:
The backup system is segmented into multiple independent components: media servers at different geographic locations, and relay servers distributed across the network. Each media server manages its own connections and backup operations independently, eliminating the need for a single complex centralized controller to manage all servers. This segmentation allows the system to scale efficiently as distributed applications grow.
2Quantity of substance
If multiple backup tapes or media are required for large distributed applications, then all data can be backed up, but the management of data across multiple servers and storage media becomes impractical
Solution Approach 1:
Relay servers automatically perform connection establishment, data transfer, and coordination with media servers without requiring manual intervention. The system self-manages the complexity of distributing data across multiple backup media by having relay servers autonomously handle the coordination, making the system easier to operate despite handling large data volumes across multiple servers.
3Productivity
If a relay server architecture is used for distributed applications, then backup data can be transmitted, but the technique becomes impractical as distributed applications grow in size and geographic distribution
Solution Approach 1:
The relay server functionality is segmented and distributed across multiple independent relay servers, each managing a portion of the data transfer operations. This eliminates the single point of complexity and allows the system to maintain high data throughput while scaling to larger distributed applications across multiple geographic locations.
Solution Approach 2:
Relay servers are designed with multi-functionality, handling both data transfer and connection management tasks. This universal design reduces the need for specialized components and simplifies the overall system architecture, making it more practical for large-scale distributed applications.
Data Source
AI summary
Backup and restore operations in a distributed data system are described. In one embodiment of a backup operation, a computing system includes a media server, and two or more distributed data sources. The media server is configured to establish a first connection with an agent executing on the first data source and receives control information specifying one or more additional data sources to contact to continue the backup operation. Agents on the additional data sources pass control information and/or backup data via each respective connection until the backup operation is complete. In some embodiments, two or more concurrent connections may be established between the media server and distributed data sources. In this manner, a backup operation may be performed for distributed data using control information supplied by the distributed data sources.