Data Copying via Region-Specific I/O Counters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data storage systems face performance issues when copying data due to the need to maintain lists of in-progress I/O requests, which does not scale well and hinders performance.
Innovation Solution
A method that involves performing a first operation to copy data from a source area, determining the count of pending I/O requests in a second region, and only proceeding with a second operation to copy data from that region once the count reaches zero, while configuring a new region to record pending I/O requests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a list of in-progress I/O requests is maintained to ensure data copying consistency, then data integrity is improved, but system scalability and performance deteriorate
Solution Approach 1:
The patent divides the source storage area into multiple regions and maintains separate counters for each region instead of a single global list. This segmentation allows the system to track I/O requests at a finer granularity, improving scalability while maintaining data integrity through region-specific tracking.
Solution Approach 2:
The patent extracts the I/O request tracking mechanism from a centralized list structure and distributes it across multiple region-specific counters. This extraction eliminates the bottleneck of maintaining a single global list, thereby improving system scalability and performance.
2Reliability
If a list of in-progress I/O requests is maintained to prevent data inconsistency, then copying reliability is improved, but processing speed deteriorates
Solution Approach 1:
By segmenting the tracking mechanism into region-specific counters, the system can independently manage and process I/O requests for each region without blocking other regions. This parallel processing capability maintains copying reliability while significantly improving overall processing speed.
Solution Approach 2:
The patent initializes counters for each region before copying operations begin, preparing the tracking infrastructure in advance. This preliminary action eliminates the need for dynamic list management during copying operations, thereby improving processing speed while maintaining reliability.
3Reliability
If a centralized list tracks all I/O requests across the source area, then data consistency is ensured, but system performance deteriorates
Solution Approach 1:
The patent replaces the centralized list with distributed region-specific counters, allowing multiple regions to be tracked and processed simultaneously. This segmentation maintains data consistency within each region while improving overall system performance through parallel operation.
Solution Approach 2:
The patent transitions from a single-dimensional centralized tracking approach to a multi-dimensional distributed counter system organized by regions. This dimensional change enables concurrent tracking of multiple I/O operations across different regions, thereby improving system performance while maintaining data consistency.
Data Source
AI summary
There are disclosed techniques for use in managing data copying. In one embodiment, there is disclosed a method including a number of steps. The method comprises performing a first operation to copy data from a first region of a source area. The method also comprises determining a count relating to pending I/O requests in connection with a second region of the source area upon completion of the first operation. And, in response to determining that the count relating to pending I/O requests in connection with the second region of the source area equals zero, the method further comprises performing a second operation to copy data from the second region of the source area and configuring a new region in the source area to record a count relating to pending I/O requests in connection with the source area.