Token Creation Latency Adaptation in Storage Replication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed storage systems face challenges in efficiently managing token creation and snapshot operations across source and target systems, particularly in balancing I/O latency limits and resource utilization during token-based copy operations.
Innovation Solution
The method involves estimating I/O latency limits and token creation times to determine whether to create a snapshot pair or a single snapshot, allowing for dynamic adjustment of token creation processes based on environmental feedback, thereby optimizing resource expenditure and minimizing latency-sensitive application jitter.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a snapshot pair is created for token creation, then data consistency and reliability are improved, but I/O latency and operation time increase
Solution Approach 1:
The system dynamically adjusts snapshot creation strategy based on real-time conditions. When I/O latency is high or the file system is busy, the system creates only a single snapshot at the source system. When conditions permit, it creates a snapshot pair with synchronous replication to the target system. This dynamic adaptation resolves the contradiction by making reliability enhancement conditional rather than absolute.
Solution Approach 2:
The system changes the replication parameter (synchronous vs. asynchronous) based on system state. The token creation process monitors I/O latency and file system busyness, then adjusts whether to perform synchronous replication to the target system. This parameter change allows the system to balance between data consistency (reliability) and operation speed (latency).
2Reliability
If synchronous replication is performed to create a snapshot pair, then data consistency between source and target systems is improved, but resource expenditure and system load increase
Solution Approach 1:
The system performs partial replication action based on needs. Instead of always creating full snapshot pairs with synchronous replication, it selectively creates only single snapshots when the target system is not immediately needed. This partial action reduces resource expenditure while maintaining data consistency when required.
Solution Approach 2:
The replication behavior is made dynamic rather than static. The system monitors target system availability and I/O conditions, then adjusts whether to perform synchronous replication. This dynamic approach optimizes resource usage by performing expensive synchronous replication only when necessary for data consistency.
Data Source
AI summary
A method is provided, comprising: receiving, from a host device, a request to create a token that represents one or more data items that are stored in a first volume, the first volume being stored in a source system; estimating an I/O latency limit for the first volume; estimating a token creation time for the token; comparing the token creation time to the I/O latency limit; when the token creation time matches the I/O latency limit: creating the token, creating a snapshot pair, mapping the token to the snapshot pair, and returning the token to the host device; when the token creation time does not match the I/O latency limit: creating the token, creating a first snapshot of the first volume at the source system, mapping the first snapshot to the token, and returning the token to the host device.