Memory Controller Partial-Array Refresh Timing Using Actual Row Counts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As DRAM capacities increase, the fixed number of refresh commands over a fixed time period becomes inefficient, leading to varying numbers of rows being refreshed in response to a single command due to segment masking in power saving modes.
Innovation Solution
A memory controller monitors segment masking and multi-row refresh schemes to determine the actual number of rows to be refreshed, adjusting the waiting period before issuing subsequent commands based on this information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the memory controller uses a fixed refresh cycle timing for all segments, then the refresh operation is simple to implement, but the memory system efficiency decreases when segments are masked in power saving modes
Solution Approach 1:
The patent implements dynamic refresh cycle timing adjustment based on the actual number of rows to be refreshed. The memory controller calculates and applies different tRFC values depending on whether segments are masked and how many rows remain to be refreshed, transitioning from a static fixed timing approach to a dynamic adaptive approach that optimizes performance for each refresh operation.
Solution Approach 2:
The patent changes the refresh cycle timing parameter (tRFC) based on the masking configuration and row refresh scheme. By adjusting this critical timing parameter according to the actual refresh workload, the system achieves higher efficiency when segments are masked while maintaining reliability, without requiring complete redesign of the refresh control logic.
2Reliability
If the memory controller waits for the full specified tRFC time after each refresh command, then data reliability is ensured, but the waiting time increases when fewer rows are actually refreshed due to masking
Solution Approach 1:
The patent applies partial refresh action by adjusting the wait time proportionally to the actual number of rows being refreshed. When masking reduces the number of rows to be refreshed, the controller waits for only that partial duration rather than the full tRFC, achieving the minimum necessary wait time for reliability while reducing unnecessary waiting time and improving overall system efficiency.
Solution Approach 2:
The memory controller performs preliminary calculation of the actual number of rows to be refreshed by monitoring segment masking configuration and the row refresh scheme before issuing the refresh command. This preliminary action enables the controller to determine the appropriate reduced wait time in advance, ensuring reliability is maintained while minimizing unnecessary waiting time.
3Productivity
If the memory controller issues commands at fixed intervals, then the command timing is simple to manage, but the efficiency decreases when the actual number of rows refreshed varies due to masking
Solution Approach 1:
The patent implements feedback mechanisms where the memory controller continuously monitors the segment masking configuration and row refresh scheme, uses this information to calculate the actual number of rows to be refreshed, and adjusts the subsequent command timing accordingly. This closed-loop feedback approach enables adaptive command scheduling that optimizes throughput while managing timing control complexity through systematic monitoring and calculation.
Solution Approach 2:
The patent transitions from static fixed-interval command timing to dynamic adaptive timing based on actual refresh workload. The controller adjusts command intervals dynamically according to the number of rows being refreshed, allowing higher command throughput when masking reduces refresh workload while maintaining proper timing management through systematic adjustment rules.
Data Source
AI summary
A memory controller combines information about which memory component segments are not being refreshed with the information about which rows are going to be refreshed next, to determine, for the current refresh command, the total number of rows that are going to be refreshed. Based on this total number of rows, the memory controller selects how long to wait after the refresh command before issuing a next subsequent command. When the combination of masked segments and the refresh scheme results in less than the ‘nominal’ number of rows typically refreshed in response to a single refresh command, the waiting period before the next command (e.g., non-refresh command) is issued may be reduced from the ‘nominal’ minimum time period, thereby allowing the next command to be issued earlier.