Cross-Data-Center Storage Failover Using I/O Statistics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In a storage system operating across multiple data centers, communication failures lead to increased billing and response times due to inter-data center communication, which conventional methods fail to address effectively.
Innovation Solution
A computer system with a storage control method that includes a tie breaker and I/O monitor to manage communication failures by switching data processing to minimize inter-data center communication, using statistical information to determine which storage controllers to stop and reroute I/O operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If failure control is performed by stopping a storage node when communication failure occurs between storage controllers in different data centers, then system reliability is improved, but inter-data center communication volume increases causing billing and response time issues
Solution Approach 1:
The I/O monitor collects communication characteristics and generates statistical information about I/O operations before failure control is executed. This preliminary data collection enables the tie breaker to make informed decisions about which storage controller to stop, predicting which action will minimize inter-data center communication volume and avoid unnecessary billing and response time increases
Solution Approach 2:
The system continuously monitors communication characteristics between storage controllers and applications, using this feedback information to dynamically determine failure control actions. The statistical information about communication patterns provides real-time feedback that guides the tie breaker's decision-making process to optimize the balance between reliability and communication cost
2Stability of the object's composition
If storage controllers in different data centers form pairs for failover, then system availability is improved, but communication latency increases due to inter-data center communication
Solution Approach 1:
Statistical information about communication characteristics is collected and analyzed before failure control is needed. This preliminary analysis identifies which storage controller in a paired configuration has communication patterns that will minimize latency impact, allowing the system to pre-determine optimal failover targets that reduce communication latency while maintaining availability
Solution Approach 2:
The tie breaker changes the operational state parameter of storage controllers based on statistical communication data. By selecting which controller to stop or activate based on historical communication patterns, the system dynamically adjusts its configuration to minimize inter-data center communication latency while preserving failover capability for system availability
3Device complexity
If conventional failure control methods are used without considering communication characteristics, then simplicity is maintained, but billing and response time increase due to unnecessary inter-data center communication
Solution Approach 1:
The I/O monitor acts as an intermediary component that collects and analyzes communication characteristics between storage controllers and applications. This intermediary provides statistical information to the tie breaker, enabling intelligent failure control decisions that reduce billing and response time without requiring complex manual configuration or overwhelming system complexity
Solution Approach 2:
The system performs self-service by automatically monitoring its own communication patterns and using this information to make autonomous failure control decisions. The tie breaker uses statistical data from the I/O monitor to independently determine which storage controller to stop, eliminating the need for external intervention while reducing inter-data center communication costs
Data Source
AI summary
In a computer system, storage controllers disposed in different data centers form a pair via a communication path between the data centers. When a communication failure occurs in the communication path between the paired storage controllers, a tie breaker takes over the data input/output from one of the storage controllers forming the pair to the other storage controller based on the statistical information of the communication characteristics of the storage controller generated by the I/O monitor, and determines failure control to stop the storage node having the one storage controller, and the storage cluster controller executes the failure control.