Data Move Engine for Transparent Non-Volatile Memory Migration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-volatile memory systems face issues such as memory wear-out and latency, leading to data corruption and delayed access.
Innovation Solution
A data move engine is introduced to relocate data from a non-volatile memory experiencing wear-out or latency issues, allowing immediate access to the data during the move by tracking the status of data lines and redirecting access to either the source or target address, thus preventing data corruption and reducing latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of stationary object
If data is stored in non-volatile memory, then data persistence is improved, but memory wear-out and latency issues occur
Solution Approach 1:
The system proactively moves data blocks before wear-out occurs by monitoring write counts and latency metrics. The data move engine initiates relocation operations in advance, transferring data from aging non-volatile memory locations to fresh locations or different memory types, preventing wear-out related failures before they happen.
Solution Approach 2:
A data move engine is introduced as an intermediary component between the processor and non-volatile memory. This engine transparently intercepts memory access requests, determines whether data has been moved, and redirects accesses to the correct location (source or target), thereby managing wear-out issues without affecting application transparency.
2Loss of time
If data is moved from non-volatile memory, then latency is reduced, but data access continuity may be affected
Solution Approach 1:
The system maintains backward pointers to source address data blocks even after moving data to target addresses. This preliminary preservation of source data allows the system to handle access requests that arrive during the transition period, ensuring continuity while enabling latency optimization for future accesses.
Solution Approach 2:
The data move engine acts as an intelligent intermediary that tracks moved data blocks and transparently redirects access requests. When data is moved, the engine updates internal mappings so that both source and target addresses can be accessed appropriately, maintaining access continuity while achieving latency reduction through optimized data placement.
3Speed
If data move engine redirects access to target address, then access speed is improved, but system complexity increases
Solution Approach 1:
The data move engine is integrated within the existing memory controller architecture, serving as an intermediary layer that adds redirection functionality without requiring separate hardware components. It maintains access speed by intercepting and redirecting memory requests at the controller level, while the complexity is managed through systematic tracking of moved data blocks using organized data structures.
Solution Approach 2:
The memory controller is enhanced with multi-functionality to include data move engine capabilities, address tracking, and transparent redirection. By consolidating these functions within a single controller unit rather than adding separate systems, the patent improves access speed while minimizing the increase in overall system complexity.
Data Source
AI summary
Examples disclosed herein provide moving a block of data between a source address and a target address. The examples disclose initiating a data move engine to move the block of data from the source address to the target address. Additionally, the examples disclose moving the block of data from the source address to the target address in a manner which allows a processor to concurrently access the block of data during the move.