Data-less Writes for Instantaneous Storage Restore
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for restore operations in storage systems are time-consuming and wasteful of resources, as they require reading data from snapshots and writing it back to the primary LU, especially when numerous changes have occurred since the last snapshot, leading to inefficiencies in data storage and management.
Innovation Solution
The implementation of data-less writes, where a snapback driver identifies changed regions and sends data-less write requests down the device stack, allowing for instantaneous restore operations without actual data payload, thereby optimizing the communication of changed data regions and reducing the need for replication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional restore operations read data from snapshots and write it back to the primary LU, then data restoration is achieved, but the operation becomes time-consuming and resource-wasteful especially when numerous changes have occurred since the last snapshot
Solution Approach 1:
The patent extracts only the changed regions from the snapshot data rather than copying all data. The system identifies and processes only the specific data blocks that have changed since the last snapshot, eliminating the need to read and write unchanged data, thus reducing restore time while maintaining data restoration completeness
Solution Approach 2:
The patent applies partial action by performing restore operations only on the necessary portions of data (changed regions) rather than the entire dataset. This selective approach reduces the overall operation time and resource consumption while ensuring that all necessary data is restored
2Reliability
If conventional restore operations write all snapshot data back to the primary LU, then data restoration is ensured, but storage system resources are wasted
Solution Approach 1:
The system extracts and processes only the changed data regions rather than all snapshot data. By identifying modified blocks and processing only those, the patent reduces I/O operations, CPU utilization, and storage system resource consumption while ensuring complete data restoration
Solution Approach 2:
The patent applies partial action by limiting restore operations to only the necessary data portions. This selective processing reduces the energy and resource consumption associated with reading, writing, and managing large volumes of unchanged data while maintaining restoration reliability
3Productivity
If data-less writes are used to communicate changed regions, then replication efficiency is improved, but the device stack must be optimized to handle data-less writes without actual data payload
Solution Approach 1:
The patent introduces data-less writes as an intermediary mechanism to communicate changed region information through the device stack. This intermediary approach allows the system to convey metadata about data changes without transferring actual data payloads, improving replication efficiency while requiring the device stack to be optimized for handling such specialized write operations
Data Source
AI summary
A method is used in managing restore operations in storage systems. A state of a logical object is retrieved and a second logical object is placed in a pre-step mode. An instantaneous restore operation between the first logical object and the second logical object is performed. Drivers that are associated with the instantaneous restore and utilize data-less writes are identified. A granularity value representative of a size at which the identified driver tracks changes is determined. Changed data extents in the first logical object are identified. Data-less writes are sent at the lowest determined granularity to any identified extents in the first logical object.