Row Hammer Detection Using Space Saving Algorithm
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional row hammer detection techniques in DRAM devices are either impractical due to high resource requirements or imperfect, leading to potential data corruption and performance degradation, especially in hyperscale datacenters where malicious attacks can exploit these weaknesses.
Innovation Solution
A deterministic row hammer detector using a modified 'space saving' algorithm that maintains a limited number of counters based on the maximum potential attacks per unit time, rather than the total memory capacity, ensuring perfect tracking without excessive energy or space consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional row hammer detection techniques are implemented, then detection coverage is improved, but resource consumption (energy and space) increases excessively
Solution Approach 1:
The patent segments the detection task by dividing the counter array into multiple banks, where each bank monitors a subset of rows. This segmentation allows the system to track row hammer attacks across the entire memory space while using limited resources per bank, resolving the contradiction between comprehensive detection coverage and energy consumption.
Solution Approach 2:
The patent implements partial action by maintaining counters for only the most frequently accessed rows (hot rows) rather than all rows in memory. The counter array tracks row activation patterns and identifies potential aggressor rows, applying detection resources selectively to high-risk areas, thus achieving effective detection with reduced energy and space consumption.
2Reliability
If conventional row hammer detection techniques are implemented, then detection coverage is improved, but device complexity increases
Solution Approach 1:
The detector is segmented into multiple independent banks, each handling a portion of the monitoring task. This modular structure reduces the complexity of any single detection unit while maintaining comprehensive coverage through the collective operation of all banks, making the overall system more manageable and less complex.
Solution Approach 2:
The system applies partial action by focusing detection efforts on hot rows that are most susceptible to row hammer attacks. By using the counter array to identify and monitor only these high-risk rows rather than all rows uniformly, the patent reduces detector complexity while maintaining effective detection coverage.
3Reliability
If perfect row hammer detection is achieved, then data corruption prevention is improved, but resource requirements become impractical
Solution Approach 1:
The patent implements partial action by maintaining counters only for hot rows that are actively being monitored, rather than allocating counter resources to every row in memory. The counter array dynamically tracks row activation patterns and identifies potential aggressor rows, providing practical protection against data corruption with a limited number of memory resources.
Solution Approach 2:
The system changes the parameter of monitoring scope by dynamically adjusting which rows are tracked based on their access patterns. Rows that exhibit high activation frequencies are identified as hot rows and allocated counter resources, while inactive rows are excluded from monitoring. This parameter change enables practical resource usage while maintaining data corruption prevention for the most vulnerable rows.
Data Source
AI summary
Practical, energy-efficient, and area-efficient, mitigation of errors in a memory media device that are caused by row hammer attacks and the like is described. The detection of errors is deterministically performed while maintaining, in an SRAM, a number of row access counters that is smaller than the total number of rows protected in the memory media device. The mitigation may be implemented on a per-bank basis. The memory media device may be DRAM.