Storage System Update Frequency Based Replication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data replication between a primary site and a backup site often causes network congestion due to frequent or bulk data transmission, which can lead to communication bottlenecks.
Innovation Solution
A storage system that determines a replication policy based on the update frequency of data, either transmitting data each time it is updated or at predetermined periods, to manage data replication efficiently and avoid simultaneous large data transfers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data replication is performed immediately upon every update in the primary site, then data synchronization reliability is improved, but network communication bandwidth is overwhelmed causing congestion
Solution Approach 1:
The patent implements periodic action by performing data replication at predetermined time intervals rather than immediately upon every update. The control unit schedules replication tasks to execute at specific periods, transforming the continuous replication process into discrete periodic operations. This reduces the frequency of replication operations while maintaining acceptable data synchronization, thereby preventing network bandwidth exhaustion and communication congestion between primary and backup sites.
2Reliability
If data replication is performed all at once at a specific point in time, then data consistency is improved, but communication bottleneck is caused due to simultaneous large data transmission
Solution Approach 1:
The patent applies segmentation by dividing the replication process into multiple discrete tasks distributed across different time periods. Instead of executing all replication operations simultaneously at one point in time, the control unit segments the replication workload into separate tasks that are executed at different scheduled intervals. This distributes the data transmission load over time, preventing communication bottlenecks while maintaining data consistency through systematic replication of updated data blocks.
3Reliability
If frequent replication is performed to maintain real-time synchronization, then data freshness is improved, but network congestion occurs due to high transmission frequency
Solution Approach 1:
The patent implements partial action by selectively replicating only the necessary portions of updated data rather than performing complete replication of all data blocks. The control unit identifies specific data blocks that have been updated and schedules replication tasks for these partial updates at predetermined intervals. This partial replication approach maintains data freshness for critical information while significantly reducing the overall volume of data transmitted over the network, thereby preventing congestion.
Data Source
AI summary
A storage system includes a storage unit and a control unit. Data is stored in the storage unit. The control unit performs replication of data that is updated, based on the frequency with which the data is updated. The control unit may include a frequency detection unit, a policy management unit, and a replication performance unit. The frequency detection unit detects the update frequency with which the data is updated, by detecting the number of times that the data is updated. The policy management unit determines a replication policy for the data based on the update frequency with which the data is updated. The replication performance unit performs replication of the updated data based on the replication policy.