Cyclic Point-in-Time Copy Architecture with Data Deduplication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data replication methods, such as IBM's Flash Copy, face performance issues in cascaded or multi-target configurations due to increased data duplication and I/O delays as the number of volumes grows, leading to slow write operations.
Innovation Solution
A method is introduced to perform writes in cyclic point-in-time-copy architectures by determining if a volume has a child volume with a target bit map, finding a higher source volume if needed, and copying data directly from the higher source volume to the child volume, thereby bypassing intermediate targets and reducing the number of copies required.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the number of volumes in cascaded or multi-target configuration increases, then the coverage and functionality of the replication architecture is improved, but the I/O performance and write speed deteriorate due to increased data duplication and copying operations
Solution Approach 1:
The patent combines multiple target volumes into a single logical target group that shares a common data structure and copying mechanism. Instead of independently copying data to each target volume, the system merges the copying operations so that data is copied once to a shared structure, and all targets in the group can access this shared data, thereby maintaining functionality while reducing redundant copying operations
Solution Approach 2:
The patent creates a universal data structure that serves multiple target volumes simultaneously. This data structure acts as a shared repository that can satisfy read requests for any target in the group, eliminating the need for separate data copies for each target and improving write performance while maintaining system versatility
2Adaptability or versatility
If the number of volumes in cascaded configuration increases, then the replication coverage is improved, but the write operation speed deteriorates due to sequential copying through multiple volumes
Solution Approach 1:
The patent segments the cascaded volume structure into independent parallel paths by introducing a shared data structure that decouples the sequential dependency. Instead of data flowing sequentially through each cascade volume, the system segments the operation so that data is copied once to the shared structure, and all cascade targets can independently access this data, eliminating the sequential bottleneck
Solution Approach 2:
The patent introduces a shared data structure as an intermediary between the source volume and multiple target volumes in the cascade. This intermediary receives data from the source once and serves all downstream targets, replacing the traditional sequential copy chain and dramatically improving write speed while maintaining full replication coverage
3Adaptability or versatility
If the number of target volumes in multi-target configuration increases, then the system can serve more clients or functions, but data duplication increases and write performance deteriorates
Solution Approach 1:
The patent merges the data storage function for multiple target volumes into a single shared data structure. Instead of maintaining separate data copies for each target volume, the system combines them into one shared structure that all targets reference, thereby serving more clients and functions while eliminating data duplication
Solution Approach 2:
The patent uses intelligent copying that determines whether data already exists in the shared data structure before copying. This selective copying mechanism avoids duplicating data that is already present, allowing the system to serve multiple targets efficiently without unnecessary data replication
Data Source
AI summary
A method for performing a write to a volume x in a cyclic point-in-time-copy architecture is described. In one embodiment, such a method includes determining whether the volume x has a child volume. The method then determines whether the target bit maps (TBMs) of both the volume x and the child volume are set. If the TBMs are set, the method finds a higher source (HS) volume from which to copy the desired data to the child volume. Once the HS volume is found, the method determines whether the HS volume and the child volume are the same volume. If the HS volume and the child volume are not the same volume, the method copies the data from the HS volume to the child volume. The method then performs the write to the volume x.