Parallel Computer Recovery via Multicast Data Fragment Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The recovery of multiple computer systems in a network is time-consuming due to the need to transfer large amounts of data, especially in disaster situations where network bandwidth is limited, leading to prolonged recovery times.
Innovation Solution
Implementing a parallel recovery method that uses data fragment multicast streams and hash queues to prioritize and efficiently transmit duplicate data fragments across the network, leveraging multicast capabilities to send common data blocks only once to multiple systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If traditional sequential recovery methods are used to transfer backup data to multiple computer systems, then data transfer completeness is ensured, but recovery time increases significantly
Solution Approach 1:
The backup data is divided into multiple data fragments that can be transmitted independently. The storage server segments the backup data and transmits these fragments to multiple computer systems in parallel, enabling simultaneous recovery operations across different systems without sequential dependency.
Solution Approach 2:
Multiple recovery operations are merged into a single coordinated process. The storage server consolidates requests from multiple computer systems and uses multicast streaming to deliver data fragments to multiple systems simultaneously, combining what would otherwise be separate sequential transfers into one unified operation.
2Productivity
If data is transmitted to multiple computer systems simultaneously, then recovery speed increases, but network bandwidth consumption increases
Solution Approach 1:
The patent merges data transmission to multiple computer systems into a single multicast stream. Instead of sending data separately to each system (which would multiply bandwidth consumption), the storage server transmits data fragments once to all systems that need them through multicast, significantly reducing total network bandwidth usage while maintaining parallel recovery speed.
Solution Approach 2:
The multicast stream serves multiple computer systems simultaneously with a single transmission. The data fragment multicast stream is designed to be universally received by all subscribed computer systems, making the transmission serve multiple functions at once rather than requiring separate dedicated transmissions for each system.
3Reliability
If all data fragments are transmitted to all computer systems, then data completeness is ensured, but redundant data transfers increase
Solution Approach 1:
The storage server receives status information from computer systems indicating which data fragments have already been received or are not needed. Based on this feedback, the storage server adjusts its transmission strategy to send only the necessary data fragments to each system, avoiding redundant transfers while ensuring completeness for systems that do receive data.
Solution Approach 2:
Each computer system receives only the data fragments it actually needs rather than all fragments uniformly. The system dynamically determines its local requirements and the storage server tailors the transmission accordingly, ensuring data completeness for each system's specific needs while minimizing redundant transfers across the network.
Data Source
AI summary
Systems and methods for parallel computer system recovery are described. The method may include receiving requests for backup data at a storage server from a plurality of computer systems to be recovered. The method may further include determining a next data fragment of the backup data to be transmitted from the storage server to the plurality of computer systems based on an order of priority. The method may also include transmitting the next data fragment from the storage server to the plurality of computer systems via a data fragment multicast stream.