DDR Memory Data Valid Window Re-determination Using Boundary Information
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
DDR memory systems require periodic training to maintain alignment between clock strobe and data edges, which can cause significant delays and responsiveness issues during system operation due to drift over time, necessitating a more efficient training approach.
Innovation Solution
The solution involves leveraging the information of the new leading edge to speed up the identification of the new trailing edge or vice versa, using boundary information to determine the new data valid window, thereby reducing the computational resources and time required for periodic re-training.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If periodic training is performed to maintain alignment between clock strobe and data edges, then data integrity is maintained, but system responsiveness deteriorates due to training delays
Solution Approach 1:
The patent performs preliminary training actions by identifying new data valid window boundaries using boundary information from previous training. The system proactively determines new boundaries before they are needed for actual data operations, so when training is required, the system can quickly apply the pre-determined boundaries without extensive re-training delays.
Solution Approach 2:
Instead of performing complete iterative training to fully re-determine boundaries, the patent uses partial action by leveraging existing boundary information and only determining what is necessary - the new boundaries of the data valid window. This partial approach sufficient to maintain alignment while reducing the time and resources needed compared to full re-training.
2Measurement precision
If iterative training steps are taken to optimize clock strobe and data settings, then alignment precision is improved, but training time increases significantly
Solution Approach 1:
The system performs preliminary determination of new data valid window boundaries using boundary information from previous training sessions. By proactively identifying new boundaries before they are critically needed, the system maintains alignment precision while avoiding the time-consuming iterative process at moments when responsiveness is most important.
Solution Approach 2:
The patent uses boundary information from previous training as a template or copy to guide the determination of new boundaries. Instead of starting from scratch with full iterative training, the system copies the structure and approach of previous successful training, adapting it to current conditions, which significantly reduces the time required while maintaining precision.
3Measurement precision
If full iterative training is performed to account for drift, then alignment accuracy is maintained, but system productivity decreases due to operation stalling
Solution Approach 1:
The system performs preliminary identification of new data valid window boundaries using boundary information from previous training. By determining new boundaries in advance and before they are needed for actual data operations, the system maintains alignment accuracy while minimizing interruptions to productivity and preventing application stalling.
Solution Approach 2:
The patent applies partial action by using boundary information from previous training and only determining the new boundaries that are necessary to maintain alignment. This partial approach is sufficient to maintain alignment accuracy while avoiding the excessive time and resource consumption of full iterative training, thereby preserving system productivity.
Data Source
AI summary
Disclosed is an improved approach to implement training for memory technologies, where a data valid window is re-determined using boundary information for a new data valid window. The information obtained for the new location of the first edge is used to minimize the computational resources required to identify the location of the second edge. This greatly improves the efficiency of the process to perform the re-training.