CAS to Non-CAS Data Migration Using Hash-to-Address Table Scanning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data migration techniques between storage systems are inefficient, particularly in Content Addressable Storage (CAS) systems, as they often result in redundant data transfers and resource wastage due to the lack of deduplication awareness, leading to longer migration times and higher resource consumption.
Innovation Solution
A method that involves invalidating data in the target volume, scanning the hash-to-address table, and pushing information from the hash-to-address table to the target, using write commands with payload and address lists, and dynamically building the hash-to-address table using a prefix to optimize data migration by only sending non-zero pages and utilizing deduplication efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional data migration techniques are used between storage systems, then data can be transferred from source to target, but redundant data transfers and zero-page transmissions occur leading to resource wastage and longer migration times
Solution Approach 1:
The system performs preliminary actions by scanning the hash-to-address table before migration to identify all data pages and their locations. This advance preparation enables the migration process to directly transfer only necessary data without redundant reads or zero-page transmissions, resolving the contradiction between migration speed and resource consumption
Solution Approach 2:
The invention extracts only the essential information from the hash-to-address table (data page mappings and non-zero page identification) and transfers exclusively this necessary data from source to target. By filtering out redundant and zero pages before transfer, the system achieves both faster migration and reduced resource consumption
2Reliability
If all data pages are transferred from source to target without deduplication awareness, then complete data migration is achieved, but redundant data transfers occur consuming excessive resources
Solution Approach 1:
The system creates a virtual copy of the hash-to-address table structure in the target system and populates it with mapping information during migration. This allows the target to reconstruct the deduplication metadata without physically transferring all source data pages, ensuring data completeness while minimizing actual data transfer volume
Solution Approach 2:
The source system performs preliminary scanning of the hash-to-address table to identify which data pages need to be transferred and which are duplicates or zero pages. This advance identification ensures that only necessary data is transferred, maintaining data completeness while reducing transfer volume
Data Source
AI summary
A method, computer program product, and computer system for invalidating, by a target, all data in a first volume. A source may scan a hash to address table for a second volume to be migrated to the first volume. Data in the second volume may be migrated to be stored in the first volume by pushing information from the hash to address table to the target.