Data Replication via Feature Information Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data synchronization across distributed data centers often faces network congestion issues, leading to unacceptable transmission latency, which conventional solutions like increasing bandwidth cannot efficiently address due to peak usage inefficiencies.
Innovation Solution
The technique separates data copies from their feature information and uses prioritized messages to transmit feature information first, determining if the data copy is qualified for transmission based on attributes like update information and location, prioritizing transmission paths to reduce latency and bandwidth usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data synchronization is performed across distributed data centers, then data consistency and disaster tolerance are improved, but network congestion occurs leading to high transmission latency
Solution Approach 1:
The patent segments data transmission into two distinct phases: first transmitting feature information (metadata, update indicators, version numbers) about the data, and then conditionally transmitting the actual data copies. This segmentation allows the system to quickly determine whether transmission is necessary without being blocked by large data volumes, thereby reducing transmission latency while maintaining data consistency across data centers.
Solution Approach 2:
The patent performs preliminary transmission of feature information before actual data transmission. By sending metadata, update indicators, and version numbers first, the system can pre-assess whether data synchronization is needed. This preliminary action enables informed decisions about subsequent data transmission, avoiding unnecessary transmissions and reducing overall latency.
2Loss of time
If bandwidth of cross-site links is increased for peak usage, then transmission latency is reduced, but cost and resource waste increase
Solution Approach 1:
The patent applies partial action by transmitting only feature information (a small fraction of the complete data) during synchronization operations. This partial transmission approach allows the system to achieve synchronization metadata updates without incurring the full cost of transmitting complete data copies, thereby reducing bandwidth consumption and associated costs while still maintaining acceptable performance.
Solution Approach 2:
The patent changes the transmission parameter from complete data copies to feature information (metadata, update indicators, version numbers). This parameter change significantly reduces the volume of data transmitted over cross-site links, allowing standard bandwidth resources to handle synchronization operations efficiently without requiring expensive peak-capacity bandwidth allocation.
3Reliability
If complete data copies are transmitted for synchronization, then data consistency is ensured, but network bandwidth consumption increases
Solution Approach 1:
The patent extracts feature information (metadata, update indicators, version numbers) from complete data copies and transmits only this extracted information for synchronization purposes. This extraction approach maintains data consistency by enabling verification of data state without requiring transmission of the entire data copy, thereby significantly reducing network bandwidth consumption.
Solution Approach 2:
The patent creates and transmits simplified copies (feature information) rather than complete data copies. These feature information copies contain essential synchronization metadata that enables consistency verification and decision-making about whether full data transmission is needed, reducing bandwidth usage while preserving data consistency guarantees.
Data Source
AI summary
The present techniques prioritize transmission of feature information associated with data copies in data transmission, and selectively transmit the data copies when necessary. As the size of the feature information is small compared with the data copies, and as feature information is transmitted using prioritized messages, delays caused by congestion in the network are reduced. Based on the use of prioritized messages, the present techniques provide solutions for replicating data across data centers when conducting various data operations including reading, writing, and modifying data. The present techniques also replicate data across data centers while providing consistency, access locality, and disaster tolerance.