Targeted Refresh Command for Row Hammer Mitigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current memory devices experience intermittent failures due to the 'row hammer' effect, where repeated access to a single row of memory within a refresh window causes data corruption in adjacent rows, and existing solutions either limit access or decrease refresh time, both of which impact performance.
Innovation Solution
A targeted refresh command is sent by a memory controller to address the row hammer condition, identifying and refreshing the victim row to prevent data corruption without restricting access or reducing refresh time, using additional pins or modifying existing commands to specify the address information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the number of accesses per row per refresh cycle is limited to address row hammer failures, then data corruption is prevented, but system performance is degraded
Solution Approach 1:
The patent applies local quality by implementing row-specific access counting and monitoring. Instead of uniformly limiting all row accesses, the system tracks access patterns for each individual row and applies refresh operations only to rows that exceed the threshold, thereby preventing data corruption in vulnerable rows while maintaining normal access performance in other rows.
Solution Approach 2:
The system dynamically adjusts refresh operations based on real-time access patterns. The refresh timing and target rows are determined dynamically through monitoring access counts during operation, rather than using static predetermined refresh schedules. This allows the system to adapt to varying workload patterns and apply refresh only when necessary.
2Reliability
If device dimensions are changed to improve DIBL and address row hammer, then reliability is improved, but manufacturing complexity increases
Solution Approach 1:
The patent changes operational parameters (access counting thresholds, refresh timing) rather than physical device parameters (transistor dimensions, channel length). This approach improves reliability by adjusting software-controlled parameters that monitor and manage row access patterns, avoiding the need to modify physical device geometry or manufacturing processes.
Solution Approach 2:
The memory system performs self-monitoring and self-refresh operations based on detected access patterns. The system automatically identifies rows requiring refresh and executes refresh operations without external intervention, eliminating the need for manufacturing changes while maintaining data integrity through adaptive behavior.
3Reliability
If refresh time is decreased to address row hammer, then data corruption is prevented, but system performance is degraded due to additional refresh overhead
Solution Approach 1:
The system performs partial refresh operations only on specific rows that require refresh, rather than refreshing the entire memory array. By applying refresh selectively to only those rows that have exceeded the access threshold, the system prevents data corruption while minimizing the time and overhead associated with refresh operations, thereby maintaining overall system performance.
Data Source
AI summary
A memory controller issues a targeted refresh command. A specific row of a memory device can be the target of repeated accesses. When the row is accessed repeatedly within a time threshold (also referred to as “hammered” or a “row hammer event”), physically adjacent row (a “victim” row) may experience data corruption. The memory controller receives an indication of a row hammer event, identifies the row associated with the row hammer event, and sends one or more commands to the memory device to cause the memory device to perform a targeted refresh that will refresh the victim row.