Adaptive Memory Controller Refresh Granularity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory systems face inefficiencies in performing refresh operations, as they often require either unnecessary data copying or lack flexibility in managing refresh operations across memory blocks, leading to suboptimal performance and reliability.
Innovation Solution
A memory controller that monitors read count values to determine whether to perform refresh operations in units of super blocks or single blocks, allowing for adaptive data management by copying data to reserved blocks when necessary, thereby optimizing the refresh process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If refresh operation is performed on all memory blocks in a super block, then data reliability is improved, but data copying overhead increases
Solution Approach 1:
The patent segments the super block into individual memory blocks and performs refresh operations selectively on only those blocks that need refreshing, rather than refreshing all blocks in the super block. This segmentation allows the system to maintain data reliability for necessary blocks while avoiding unnecessary data copying overhead for blocks that don't require refresh.
Solution Approach 2:
The patent applies partial action by performing refresh operations on only the necessary portion of memory blocks (those with read count values indicating need for refresh) rather than performing excessive refresh on all blocks in the super block. This reduces unnecessary data copying while maintaining adequate refresh for blocks that need it.
2Loss of time
If refresh operation is performed on individual memory blocks, then data copying overhead is reduced, but management complexity increases
Solution Approach 1:
The patent implements a feedback mechanism by monitoring read count values of memory blocks and using this information to dynamically determine which blocks require refresh operations. This feedback-driven approach automates the selection process, reducing management complexity while enabling precise control over which individual blocks undergo refresh, thereby minimizing unnecessary data copying.
3Ease of operation
If refresh operation is performed in units of super block, then operational simplicity is improved, but data copying overhead increases
Solution Approach 1:
The patent introduces dynamics by making the refresh operation granularity adaptive rather than fixed. The system dynamically adjusts between refreshing individual blocks or larger groups based on monitored read count values and system conditions. This dynamic approach maintains operational simplicity through automated decision-making while reducing data copying overhead by avoiding blanket refresh of entire super blocks when not necessary.
4Productivity
If adaptive refresh management is implemented, then refresh efficiency is improved, but device complexity increases
Solution Approach 1:
The patent implements self-service by enabling the memory controller to automatically monitor read count values and autonomously determine which memory blocks require refresh operations without external intervention. This self-service mechanism improves refresh efficiency through adaptive decision-making while managing device complexity by using straightforward monitoring and selection logic based on read count thresholds.
Data Source
AI summary
A memory system includes a memory controller. The memory controller includes: a buffer memory configured to store read count values of memory blocks and address mapping information of the memory blocks; and a central processing unit configured to monitor the read count values, and determine whether a refresh operation of target blocks, among the memory blocks, is to be performed on all memory blocks including target blocks in a super block, in a unit of the super block or on the target memory blocks in the super block in a unit of a single block, based on the monitoring result.